Despite the variety and coverage of microfinance in Indonesia, most of rural
households had no access to saving and credit services from a semiformal or
formal financial institution, even if they have a viable commercial enterprise
(Charitonenko and Iswan, 2002). Poor access to credit correlates
strongly with deficiency in income and lack of assets for collateral, generally
land ownership (Pattern and Jay, 1991; Mizuno,
1996; Johnston and Jonathan, 2008). Most of them
have to rely on informal sector, such as moneylenders, money keepers, financial
arrangements tied to land, labor and traded goods, friends and relatives and
financial self-help groups for financial intermediation. Third-party guarantees,
tied contracts and threats to future access to credit are common devices in
informal contracts (Adams and Fitchett, 1992;
There is growing interest in the importance of financial Self-Help Groups (SHG)
in developing countries since it sustain under social and geographic isolation
and reported to have a very low default rate (Biggart, 2001).
Bouman (1995) introduced two basic types of financial
SHG with savings and lending as their primary function, those with rotating
and those with nonrotating funds. The first type is widely known in the literature
as ROSCA, for Rotating Savings and Credit Association. The second type is called
as ASCRA, for Accumulating Savings and Credit Association. Both types form an
important part of all three main functions of self-help groups: security or
insurance, financial and social.
ROSCAs have become a widespread phenomenon in rural and urban economics around
the world (Ardener, 1964). A large number of Indonesia
households, most particularly including women, have a strong tradition of participating
in ROSCAs, widely known as arisan (Geertz, 1962; Varadhanarajan,
2004; Hospes, 1996; Kawagoe et
al., 1992). In rural Java, nearly every village has an arisan group
(Shigetomi, 2004). The number of arisans groups in Indonesia
is estimated to be in the millions (Martowijoyo, 2007).
Geertz (1962) believed arisan in rural areas has become
more and more important as strictly economic institutions due to the intense
need for a cash resource.
Van-den-Brink and Jean-Paul (1997) stated that ROSCA
is an association of men and women who meet at regular intervals and distribute
a lump sum of money to one of its member. Loan is automatically distributes
by rotation among ROSCA members thus generates incentive problems. The prospect
of participating in future cycles of the ROSCA and social sanctions by the other
members of the group are tools to decrease opportunistic defection (Besley
and Alec, 1996; Ardener, 1964). Enforcement is also
possible through exclusion, allocation of ranks and membership fee (Anderson
et al., 2008).
ASCRA is another endogenous financial SHG that has developed in rural Indonesia,
which known as simpan pinjam (Soemardjan, 1963; Shigetomi,
2004). ASCRAs resemble credit unions. The pooled savings are kept in custody
and accumulated for a specified time, then redistributed to members. The common
period is one year, so it can meet member demand for paying taxes, education,
a religious ceremony, or social activities (Bouman, 1995).
Loans from ASCRAs require a decision from a loan committee or the consent of
members. However, most of study on ASCRA in Indonesia focused on loan-subsidized
group by government or NGO rather than indigenous one (Seibel
and Uben, 1992; Ismawan, 2000).
There is still limited information regarding a groups characteristics and credit constraint problems of financial SHG, especially in remote areas in Indonesia, although such information can result in a better understanding of a households participation and development of financial SHG. The earlier literature of financial SHG in Indonesia subsequently addressed questions of efficiency and effect regarding ROSCA or ASCRA by ignoring factors that influence individuals to participate in the first place. By taking institutional characteristics into consideration, social-based information should be discussed, since both ROSCA and ASCRA belong to local level community associations.
The crucial assumption regarding financial SHG participant is that individuals
are credit constraint by formal institutions (Besley et
al., 1993; Handa and Claremont, 1999). In contrast,
Varadhanarajan (2004) found the opposite condition for
Indonesia case. However, there is limited information of credit constraint by
financial SHG although it able to gain and utilize information from community
to decide member creditworthiness. Zeller (1994) found
that similar with formal institutions, informal lender might consider wealth
and leverage ratio as criteria for credit rationing. Schreiner
and Geetha (1998) study found that being female and having borrowed from
other informal lenders are good predictors of being creditworthy by ROSCA or
This study will concentrate on development financial SHGs in a remote upland village in West Java since most households living in geographically isolated areas have poorer access to formal financial institutions due to collateral requirements and high transaction costs. Most of rural households in West Java are excluded from formal credits market since only 14% of land has land title. The objectives of the study are: to determine factors to promote rural household participation in ROSCAs or ASCRAs and to determine factors influencing households probability to receive credit rationing from ROSCAs or ASCRAs.
MATERIALS AND METHODS
Research area: Research was conducted in Kemang Village, which is a mountainous area in the Eastern end of the valley of the Bojong Picung sub-district of Cianjur, West Java. It lies 25 km Southeast of the district capital. The village area is 2,529 ha and is divided into 22 hamlets (kampong). Access to Kemang Village is limited since public transportation only reaches the nearest village and then continues by motorcycle or walking. Other than this main road, there is only a footpath is connecting hamlets up to higher and hilly areas.
Kemang village represents the upland area that is surrounded by the National
Forest Land (NFL). Of the total village area of 2,529 ha, 44% of it belongs
to national forest land. The land that belongs to the people of Kemang consists
of 87.8 ha rice fields or sawah (4%) and 878.6 ha dry land or pasir. Rice production
is cultivated twice a year but mainly for self consumption. Meanwhile in the
dry land area, most of the households are engaged in a huma-talun system (Mugniesyah
and Mizuno, 2007), a typical agroforestry system in the Sundanese community
in West Java in which people cultivate the land with a mixture of annual and
perennial plants. The commercialization of dry land commodities contributes
more cash income than paddy production (Tsujii, 2002).
The cultivated commodities on dry land are Banana (Musa sp.), red rice
(Oryza sativa), sugar palms (Arrena pinnata), cassava (Manihot
esculenta), pepper (Piper nigrum), Sengon (Albazia falcata),
maize (Zea mays), vegetables and fruits trees.
Increasing market demand for banana leaves for wrapping food in urban areas
increased banana leaf production by 30% per year during 1998-2003. The topographical
condition gave advantage to banana leaf quality since less wind creates fewer
cuts on the banana leaves. Banana leaf production is accessible to all households.
The middle and the landless households can cultivate the national forest land
through temporary agreement with the state forestry corporation (informal rights)
or even illegal cultivation (Inoue et al., 2003).
In addition, the patron-client relationship creates opportunities for them to
work in the harvesting and transporting of banana leaves for the upper stratum
households. The more remote the dry land is, the higher is the share obtained
by the laborers (Mugniesyah and Kozuke, 2007).
Banana leaf production delivers regular cash income since it is harvested about
eighteen times a year, using about a week each time. Farmers can choose to sell
banana leaves either to hamlet or village traders by considering the most convenient
place, kinship relation, or patron-client relationship with traders. Despite
the fact that the banana leaf marketing structure in the Kemang village is an
oligopsony (Asmarantaka et al., 2003) traders
maintain a marketing network with farmers by giving gifts, supporting social
activities and improving public facilities (roads and Islamic boarding school
Data and respondents: This study uses primary data collected from in-depth interviews of upland households in two adjacent hamlets, Beber and Cikupa. These two hamlets are in the central part of Kemang Village, where public services facilities such as a primary and secondary school, village office, etc. and economic activities are located. First, the household socioeconomic survey and interview of key informants related to ROSCA and ASCRA development were conducted in 2003 and 2005. Second, an intensive survey regarding households savings and credit access from both formal and informal financial institutions was conducted in July-August 2006 and September-October 2007. The primary data on households individual characteristics, landholding, assets, income, expenditure and households memberships in local institution were collected.
The households sample consists of 74 households or 44.8% of the total households in the two hamlets. The households were stratified into 4 strata based on two criteria: (1) ownership value of their rice field and (2) ownership value of their dry land. Stratum A has land value of more than 56 million, stratum B between 28-56 million, stratum C less than 28 million and stratum D is landless.
Households participation in financial self help group and credit constraints: A two stage analysis is needed to determine the factors influencing the development of the financial SHG in rural Indonesia: (1) identifying factors influencing rural household participation and (2) identifying factors influencing rural households faced credit constraint from financial self help group.
Model of participation in financial self help group: Households participation on ROSCA and ASCRA shows demand for credit and access to an informal credit market. ROSCA participation means creditworthiness since every member received equal credit. However, participation to ASCRA not necessarily means credit access since their credit application will be decides by groups committee.
The logit model is used to estimate factors influencing households to participate
in ROSCA or ASCRA. The dependent variables have binary values of 0 if not participate
and 1 if participate. The following equation is used to estimate the probability
of joining ASCRA or ROSCA:
Prob (Participation) = F (I, H, S, W,
||Vector of individual characteristics of households affecting participation
to financial SHG (age (+), education (+), sick days (+), being a government
employee (+), having non-steady income (+) )
||Vector of endowment in human capital (education (+), dependent (-))
||Participation in social activity (S (+) measuring membership and position
in local association)
||Vector of household assets (rice field and dry land value)
||Vector of household events that are expected to affect participation in
financial SHG (costly social events such as marriage (+))
Households characteristics, asset, type and income and social network
in local associations are considered to influence participation of financial
SHG. Type of income identified by share of agricultural income (exclude banana
leaves income), brown sugar and labor may influence decision to participate.
Households consider having non steady income if share of agricultural income
is higher than 50% of total income.
Local associations included in the indicator are womens groups, Koran
reading groups, water management groups, sports groups, youth groups, brown
sugar farmers groups, forest farmers groups, groups for education, pension groups,
teacher groups and village development groups. Social participation index estimate
by this following equation:
Social participation =Σ Household
membership in local indexassociationxpositionsxmeeting in the last three
Model of credit constraint: Credit constraint identifies directly from
the households surveys by asking whether households consider themselves
to be credit constrained by using a carefully designed credit questionnaire
(Zeller, 1994; Sawada et al.,
2006). To identify credit constrained households, first households were
asked whether they have applied for credit in the last two years. Among those
who apply for credit, households were identified as unconstrained if they received
as much credit as they requested. Meanwhile, households were considered external
credit constrained if their loan application was rejected or they could not
Second, for those who did not try to borrow, respondents were asked why they did not apply for financial SHG credit in order to further classify them. Households were identified as being unconstrained if the answer was no need for credit. Respondents were identified as being internal credit constrained if:
||They already had a large amount of debt
||They believe credit application will be refused
||They do not know anyone who could be a guarantor
||They do not know how to get credit or do not know any SHG
||They feel ashamed to ask for credit
||They are afraid to apply for credit
The logit probability model was used for both the external and internal credit constrained, since the dependent variable has a binary value. In the external credit constraint, the dependent variable has binary value of 0 if the households credit application was not rationed or rejected or 1 if it was rationed/rejected. The following equation is used to estimate the probability of households receiving an external credit constraint:
Prob (External credit constrained) = F
(I, H, S, W, E, R)
Meanwhile, the following equation is used to estimate the probability of households having internal credit constraint:
Prob (Internal credit constrained) = F
(I, H, S, W, E, R)
||Vector of individual characteristics of households affecting the committee
financial SHG decision on the loan application (age (-), education (-),
sick days (+), working as government employee (-), having non-steady income
||Vector of endowment in human capital (education (-), dependent (+)
||Participation in social activities (-)
||Vector of household assets (rice field and dry land value (-), total income
||Social event that required high expenditure (wedding celebration (-))
||Vector of repayment ability variables (ratio of debt over last years
RESULTS AND DISCUSSION
Kemang households income and demand for mutual financial institutions: Table 1 shows the income structure of respondent households according to strata. Households income calculation excludes irregular incomes such as gifts, remittances and government subsidies. The average households total income grew 1.5% from 14.8 million IDR in 2006 to 15 million in 2007. All households in Kemang village regardless their landholding conducts pluriactivity. The non-agricultural activities have a significant contribution to total income, especially for the highest stratum (stratum A) and the landless (stratum D). Meanwhile, agricultural income still plays a major role in the middle (stratum B) and the low (stratum D) stratum households total income.
Agricultural income is closely related with economic and employment opportunities
in highland farming. Income from wet rice farming, agricultural labor, palm
sugar, livestock and banana leaves production varies according by stratum. There
is tendency that larger landholding implies to higher agricultural income. Rice
farming, which produces twice a year, is mainly for self consumption. The highest
stratum produces the most.
||Average total income by stratum in Kemang Village, 2006-2007
||Households participation in ROSCA and ASCRA by stratum, 2006-2007
|Number in parentheses is participation rate (%). ROSCA in
kind is excluded due to low sustainability
Dry land delivers higher income than rice fields for all stratum households. The commercialization of banana leaves delivers regular cash income since it can be harvested 18 times per year. Another important source of income from dry land is palm sugar, especially for middle and low stratum households (B and C). Other commodities produced from dry land are peppers, red rice, peppers and fruits.
In terms of non-agricultural income, households from high and low strata conduct different types of off-farm activities. Some households from upper strata (A and B) generate income by working as government employees (teachers or pensions), banana leaf traders, or providing rice milling service. Meanwhile, the lower strata (C and D) generate income from trading (timber trading, grocery stores), industrial labor and motorcycle taxi services (ojek).
Growing economics activities after a boom of banana leaf production increase demand for saving and credit services. The geographically isolated location of Kemang village constitutes a natural barrier to access formal financial institutions. This situation encourages Kemang households to utilize their networking and develop financial SHG. Table 2 shows an interesting pattern of household participation in informal credit markets. There is a tendency for higher stratum households to have higher participation in ASCRAs. In contrast, the middle and the low stratum households have a tendency to participate in ROSCAs. However, the landless have little access to both ROSCAs and ASCRAs, since participation is possible only if they have surplus income. Especially for ROSCA, participation requires a regular cash income flow.
Household Access to ROSCA in Kemang Village: The development of ROSCA or arisan in Kemang Village is highly influenced by upland farming activities. Unlike the characteristics of ROSCA described in the earlier, the ROSCA in Kemang Village shifted into a prominent economic association. As banana leaf and brown sugar production required intensive labor work, most ROSCA groups conduct no regular meeting. All groups apply a fix order of lots or called kocok mati in the local language. The order of lots was made by member negotiation rather than a bidding auction under a managers leadership. However, it is opens to change in case one of the members experiences a sudden misfortune such as illness or death.
The managers, who are largely managers by profession, have an important role in ROSCA. They are in charge of screening potential members, collecting and distributing the fund. They are also responsible for smoothing out problems and compensating for any loan default. In return, they have the benefit of receiving the first lot and commission. This potential benefit has increased competition among managers by offering different size, duration, or commission rule. In the Beber hamlet, almost all groups have a fixed rule on commission (one or half lot payment). However, all groups in Cikupa agree to give commission based on each members willingness to pay. In both hamlets, there is no case of managers give loans to needy members by charging interest.
There is a relationship between membership and frequency. Short frequency ROSCA has more group members. Weekly ROSCA has a bigger membership than that of the monthly ROSCA (20-80 people compared to 12-21 people). Further characteristics among ROSCA groups in Kemang village are shown in Table 3.
Unlike earlier study, which highlighted a social function for meeting especially among women, only a small number of groups formed based on similarity of gender, kinship and occupation. ROSCA members are mostly close acquaintances and know each other through daily interaction, especially through religious activity in the mosque. It is not only popular among women since most ROSCA groups consist of mixed gender members. The exclusive womens groups are based on residential propinquity or common association (Koran reciting group, health group). Meanwhile men exclusively groups are based on occupation similarity (taxi driver service (ojek)).
Table 4 shows the different functions of ROSCA by. Weekly ROSCA is mostly used for smoothing consumption as a security against fluctuation of expenditure, especially for celebrating the end of Ramadan (lebaran), facing health problems, paying education fees and others. Meanwhile, monthly ROSCA has the potential to contend capital scarcity problems, since it provide credit for productive activity. It provides capital for agricultural production and land mortgage.
ASCRAs in Kemang Village: ASCRAs were first developed by a few elementary
school teachers in Kemang to provide credit and saving services.
|| Characteristics of ROSCA groups in Kemang Village, 2006-2007
|Group 1-7 are weekly ROSCA. Group 8 is a bi-weekly ROSCA.
Group 9-15 are monthly ROSCA
|| ROSCA function by frequency and stratum, 2007
the group had homogenous members, it developed into heterogeneous group after
membership become open to the Kemang community. There are two ASCRA groups currently active in Kemang Village. The first group
was started in 1983 and is named Silih Asih, which means to take care of each
others. The second group, Simpay Wargi, means bond of family and was started
in 1986. The membership is basically spread through kinship, friendship and
working relationships. Both ASCRAs groups have large memberships of more than
ASCRA is similar to a credit union. The fund accumulates over time, allowing members to take short- and long-term loans. Both groups are charged a high loan interest of 36% per year. The credit terms can be extended under negotiation with the committee. The savings and credits transactions make fund grow fast and deliver good rewards to the saver. The profit is distributed in an annual meeting: committee (15%), bookkeeping (10%) and distributed profit for member (75%). ASCRA has a social function since the profit distribution allows members to celebrate Ied (lebaran) and Independence Day (August 17).
Members usually apply for loans in the annual meeting. They can borrow more
money than their savings, which potentially causes liquidity problems to the
groups. There is no written contract and collateral requirement for a credit
transaction, even for large and long-terms credits. However, loans from ASCRA
are not automatic since they require approval from a board of loans committee.
|| Function of ASCRA credit in Kemang Village, 2007
This system may cause inequality credit access among members since the committees
give priority to their inner circle committees (relatives and teachers).
Table 5 shows the function of ASCRA credit to member applicants. ASCRA is able to deliver credit for productive activities. It can provide larger loans under long- term contracts so that members are able to mortgage and even buy dry land. However, credit is also important for consumption purposes since, a third of credits were utilized for celebrating wedding ceremonies and paying medical expenses.
Both groups have imperfect information problems in fund management. Members
have limited information about committee decision on loan distribution and safekeeping.
As the group grows bigger, the committee decides to deposit part of the fund
in a local semiformal bank to reduce safekeeping risks. Few members reported
receiving credit constraint. There is tendency of misuse of power by the committee
member by giving loan privileges to teachers and close relatives. Both groups
are having serious liquidity problem because loan default and loan term extension.
As a result, committees have to reject or only partially approved credit applications.
In addition, members unable to withdraw their savings in under short notice.
Both credit rationing and withdrawal problems thus deteriorated member trust
of the committee, especially by those who received credit constrained. Table
6 shows the decreasing number of membership in both groups over the last
|| Development of membership, savings, credit and distributed
profit of ASCRA in Kemang Village, 2000-2007
|Field survey (2005-2007)
Participation in financial self help groups and credit constraint problems: Households participation in borrowing depends on their access to financial SHG and demand for credit. A further explanation will divide households experienced in ROSCA and ASCRA. The first section will describe factors influencing households to participate in any financial groups. The second section will describe factors influencing credit constraint of Kemang households. There are two types of credit constraint, external and internal. Households are identified as external credit rationing if they receive a rejection or partial ration loan from ASCRA or ROSCA (supply side). Households were questioned about how much they asked to borrow and whether it was from financial groups approved or rationed by their lender. Meanwhile, internal credit rationing occurs when households decide not to apply for credit, since they perceive no chance for receiving any credit, so they find it not even worth trying.
Participation in ROSCA and credit constraint
Participation in ROSCA: ROSCA mainly functions as an economics institution,
providing credits for consumption and production activities. Here, we try to
determine the factors influencing households to participate in this institution.
Table 7 shows factors influencing households to participate
in ROSCA. First, the age of the household head has a significant influence on
their probability of participating in ROSCA. Households with older household
heads are more likely to join ROSCA since they presumably have more control
on household resources. This relationship is declining, however, with higher
age. The average head of households age who participates in ROSCA is 47
years. Second, households with higher household head education have a higher
probability of participating in ROSCA. They might have better knowledge to anticipate
fluctuation of income and expenditure in agricultural economies. Third, total
income has indifferent influence on households participation in ROSCA.
|| Factors influencing households participation in ROSCA
|N = 148; Chi-square = 41.1; Percentage correctly predicted
= 66.9; ***Significant at α = 1%; **Significant at α = 5%; *Significant
at α = 10%
The middle stratum household has same opportunity to participate in ROSCA
regardless of their total income value since they can select groups with suitable
frequency, management fee, or lot value. The highest stratum households might
believe ROSCA does not deliver suitable financial intermediation for them. Meanwhile,
the lowest stratum might be reluctant to participate due to a low surplus and
irregular income. Fourth, households with non-steady income, proxied by a higher
percentage income derived from agricultural activities excluding banana leaf
income, has a lower probability of participating in ROSCA. ROSCA participation
required regular cash income since each member contributes equal and regular
payment. Households having non-steady income are less likely to participate
in ROSCA to avoid loan default.
Credit constraint in ROSCA participation: In ROSCA, external credit
rationing is not taken into consideration. There is no case of households being
rejected to join ROSCA. Managers have a tendency to invite members based on
trust and collective obligation. However, managers are able to conduct partially
credit rationing by rationing the number of lots in future ROSCA circles, though
this rarely happens. Among non participants, we questioned internal reasons
not to borrow/join ROSCA.
||Factors influencing household of being internally credit constraint
|N = 148; Chi-square = 40.2; Percentage correctly predicted
= 73.6; ***Significant at α = 1%; **Significant at α = 5%; *Significant
at α = 10%
Households can be identified as credit constrained if their answers are (1) we already have enough debt and (2) we are afraid we cannot make a regular contribution.
Table 8 shows factors determine households of being credit internal constrained from ROSCA. There are three factors significant influencing households probabilities of being credit constraint. First, older households head has lower probability of being internal credit-constrained. Older households head presumably have better control over households resources. However the relationships increase with higher age. Second, of households head work as government employee are less likely of being credit constrained by ROSCA. They earned relatively steady cash income from salary so able to fulfill regular money contribution. Third, in contrast with present hypothesis, social participation has positive influence with credit constrained. Households that participate in many social activities in the community are more likely to be internal credit constrained. They were reluctant to participate in ROSCA since it becomes a strictly economic institution but utilizes social sanctions for default enforcement. Households living in a close tied community have incentives to keep their social reputation and responsibility in the community. Fourth, rice field value has indifferent influence on a households probability of being credit constrained. Rice fields deliver no cash income since rice production mainly for self consumption.
Credit access and credit rationing from ASCRA
Participation in ASCRA: As stated before, there are two developing
ASCRA groups in Kemang Village. Table 9 shows factors influencing
a households decision to participate in ASCRA. There is only one variables
significantly influence households to join ASCRA. Households with the household
head working as government employee have a higher probability of participate
in ASCRA. Historically, a teacher has higher access to join ASCRA, since it
first developed among teachers to increase their family welfare.
|| Factors influencing households participation in ASCRA
|N = 148; Chi-square = 27.5; Percentage correctly predicted
= 64.4; ***Significant at α = 1%; **Significant at α = 5%; *Significant
at α =10%; Chi-square = 27.5
Membership has increased through kinship relations by recruiting the inner
circles families of teachers. ASCRA committees, which belong to the highest
stratum households, also utilize patron client relationships to recruit members
from the lower stratum households. Nowadays ASCRA groups develop have large
and heterogeneous memberships.
Credit constraint in ASCRA: In this study, among 76 credit applications, 54 households were identified as non credit constrained and 22 households were identified as externally credit constrained (16 partially granted, 6 get rejected). Internal credit rationing was identified among non-borrower households that decided not to apply for credit since they felt no chance of receiving credit. In non-borrowing households, 12 households were identified as unconstrained and 60 households were identified as internal credit constrained.
Table 10 shows the factors influencing the probability of
receiving external and internal credit constraint from ASCRA. There are three
factors significantly influencing households probability of being externally
credit-constrained. First, households with a poor health condition, proxied
by high number of sick days, are more likely to be perceived as externally credit
constrained by ASCRA. Although, some households first, decide to join ASCRA
for insurance during poor health, unfortunately their loan applications get
rejected. The committee might consider them risky borrower since households
have to pay high medical costs under income fluctuation. Second, households
who celebrated a wedding have a lower probability of receiving credit constraint.
Wedding celebrations are only held by well-off family member show economic power
and maintaining the family reputation. The ASCRA manager may approve loan applications
since wedding celebrations potentially generate income from family and quests.
||Determinants of being externally and internally credit-constrained
by the ASCRA group (logit estimate)
|For external credit-constrained: N = 148, Chi-square = 32.4,
Percentage correct = 89.2, For internal credit-constrained: N = 148, Chi-square
= 114.1, Percentage correct = 87.8. ***Significant at α = 1%, **Significant
at α = 5%, *Significant at α = 10%
Third, households saving accounts have indifferent influence on households
probability of being credit constrained. This result supports the issue of misuse of funds by powerful committee members.
There is a tendency that family and members with strong connections to the committee
have privilege to borrow much more money than their savings account under long
term contracts. On the other hand, among big savers, widows and households with
poor health may receive credit rationing.
Meanwhile, internal credit rationing was influence only by the ratio of debt over last years income as a proxy of repayment ability. Households with a higher ratio have a higher probability of hesitance in applying for credit from ASCRA. Households might feel reluctant to apply for credit unless it generates income since they are charged a high interest rate. In addition, they perceive no chance of receiving any credit before returning all debt, so they might feel it is not even worth trying. Savings accounts and total income have indifferent influence on the probability of receiving any credit constraint. Households have the same opportunity of being credit constrained regardless of their savings. Meanwhile, total income may not be useful as a factor to consider a household internally credit constrained. Regardless of their total income, households might have the same probability of not applying for credit.
A financial SHG in a geographically isolated area can replace a limited coverage
of government agencies to provide financial services for rural households with
low income and lack of collateral. Both ROSCA and ASCRA strictly perform an
economic function and are highly influenced by upland farming activities. As
informal institutions, financial SHG can obtain more information that is only
available inside a community to judge member creditworthiness. However, a households
decision to participate in a financial SHG influences by the characteristic
of the group itself.
ROSCA is built among close acquaintances with different kinship, gender and occupational background. Banana leaf production played an important role in ROSCAs development. Despite the fact that ROSCA participation required equal and regular fund contribution, it is possible for middle and lower strata to participate in ROSCA after they have a steady income from commercialization of banana leaves. Households socially active in the community may conduct internal credit rationing by reluctant to participate in ROSCA to maintain their social reputation, since ROSCA only perform economics functions by utilizing social sanctions as enforcement tools.
ASCRA members mostly come from the highest and middle stratum households that work as government employees since it historically developed among teachers. It creates a misuse of power among committee members by giving privileges to close family and friends. In contrast with formal institutions, committees are able to obtain member health condition and information regarding wedding celebrations as a proxy for economic and social power information to decide on loan applications. Members with low repayment ability may encounter internal credit constraints.
The Author would like to extend sincere gratitude to Prof. Takenori Matsumoto for his patient guidance and valuable comment on to the draft paper. The research would not have been possible without the JSPS Core University Program between The University of Tokyo and Bogor Agricultural University: Prof. Noriaki Iwamoto, Prof. Kozuke Mizuno, Prof. Hiroshi Tsujii, Siti Sugiah M. Mugniesyah, MS, Dr. Sri Hartoyo, Dwi Rachmina, MS and Rokhani, Msi.