Agriculture Credit in Developing Economies: A Review of Relevant Literature

This paper aims to present a comprehensive review of 110 studies on agriculture credit in developing countries during 1995 to 2015. The literature has been classified and presented on the basis of time period, country of study, methodology used, issues covered, and sources of study. Agriculture credit has gained interest of policy makers and researchers in developing economies in recent years with raising concerns of issues like food security and rising population. However, the situation of small and marginal farmers is still vulnerable and they lack timely and adequate access to institutional sources of finance. Non-institutional sources of credit are still dominant in rural credit markets; while the role of micro-finance appears dubious. This study will prove helpful for policy makers and future researchers who wish to study diverse issues in rural finance in general and agriculture credit in particular.


Introduction
Agriculture is an essential economic sector of all world economies-be it developed, developing or under-developed, but it is the most important sector of a developing economy in terms of output and employment generation as compared to other sectors.(Soubbotina & Sheram, 2000).Agriculture plays a predominant role in economic development of developing economies.Developing economy is one such economy which is characterized by the presence of both rural and urban sectors and is heavily dependent on agriculture (Mylott, 2009;Fan et al., 2005).The agriculture sector not only fosters the growth process of these economies but also provides food to their ever-growing population and provides employment to larger parts of their workforce.It is the backbone of an economy which supports rest other sectors.While the manufacturing sector needs direct input from agriculture in form of raw material, service sector is indirectly dependent on agriculture.Agriculture sector is crucial for both rural and urban sectors of an economy as it generates employment opportunities in the former and provides food and raw material to the latter.Besides it crucial importance in the overall development process, farmers in developing countries are to a large extent constrained by credit.The non-availability of adequate credit when needed negatively impacts the farm output (Guirkinger & Boucher, 2008;Feder et al., 1990).The exclusion of masses from basic services of a financial system leads to significant loss in gross domestic product (GDP) of a country (Chattopadhyay, 2011).As the agriculture sector in such economies is dominated by small and marginal farmers, governments play an active role and initiate several policy measures time-to-time to improve situation of such farmers (Khandker & Koolwal, 2015).Still the majority of these farmers lack the timely access to institutional credit in adequate amounts needed in the production process.
Therefore, it becomes necessary to study the constraints which hinder the outreach of institutional credit to such vulnerable groups.Since the problems of farmers in developed countries are different from those in developing countries (Jansson et al., 2013) and given the crucial importance of agriculture sector in developing economies, this study has reviewed relevant literature on agriculture credit in emerging and developing economies.Further, the countries are categorized as "emerging and developing economies" by International Monetary Fund 's (IMF) "World Economic Outlook Report 2015".

The Conceptual Framework
Agriculture sector is a major contributor of GDP of agriculture-based economies as compared to other sectors of the economy and is a primary source of livelihood for more than half of their total workforce (Mondiale, 2008).
Credit is needed as an important indirect input among others to enhance productivity in agriculture (Sriram, 2007;Das et al., 2009).With modernization and mechanization of farming systems, farming communities require more farm investment.Since most of the farmers in developing countries are small and marginal with fragmented land holdings, they need credit for such investment.Due to lower rate of savings in these economies, the farmers lack sufficient owned-equity and hence resort to external borrowings (Chisasa & Makina, 2012).
Most of the farming households are faced with paucity of funds at their end.To fulfill their credit requirements, both institutional and non-institutional of finance are available in a developing economy (Singh et al., 2001).When credit is not available on time and at reasonable rates from institutional (formal) sources, farmers are forced to pay exorbitant rates of interest to non-institutional (informal) lenders (Reddy, 2012;Chaudhuri & Gupta, 1996).Traditionally when agriculture was mainly subsistence based, informal moneylenders used to cater to credit needs of farmers which were comparatively small.After the Green Revolution across the world which initiated tremendous changes in the cropping pattern, the credit needs of farmers have increased spontaneously; and it was during this period that institutional sources of credit emerged as major players.This was the era when subsistence cropping was replaced by cash cropping.Later on, micro-finance emerged as an effective tool of providing credit to the rural communities (Pradhan, 2013).Figure 1 shows the principal sources of credit available to farming communities of an economy.Credit from institutional sources include credit from setup of institutional framework with Apex bank of the country at the top and institutions covered under its purview including specific bodies established for agricultural development of the nation, commercial banks, co-operative banks, regional rural institutions, whereas the non-institutional sources cover credit from unorganized sector like friends, relatives, big landlords, contractors which are not part of the institutional setup.In the mid -way of institutional and non-institutional agencies, lies the semi-formal setup of micro-finance -the provision of a range of financial and non-financial services to group members based on joint liability.While the financial services of micro-finance institutions (MFI) include providing loans (generally in small amounts) to group members, insurance cover, provision of savings; the non-financial services include training and self-employment programs at an affordable cost (Morduch, 1999).
Figure 2 highlights the components covered under the scope of institutional credit.While the direct credit includes all short medium and long term loans for agriculture and allied activities to farmers with direct responsibility of repayment to the lending agency, indirect credit on the other hand includes indirect farmer benefits through subsidized farm inputs.In case of indirect credit, the farmers are under indirect repayment responsibility to the lending agency through fertilizer dealers, corporations, input supplier.
Despite its crucial importance and efforts by government, there exists shortage of agriculture credit in relation to its demand by farming communities.This unmet demand paves way for indigenous lenders as a source of finance for farmers.In this regard, this paper is an attempt to present the financing problems/constraints of farmers in developing economies as reported in literature.A comprehensive review of agricultural credit in various developing economies in the world has been presented covering a time span of twenty years during 1995 to 2015.Published and unpublished literature has been surveyed to analyze determinants, status and performance, determinants of agriculture credit in various countries included in the study.

Rationale and Scope of the Study
In the recent years, there has been a growing concern about farmer distress, productivity in the agriculture sector amidst rising concerns over food security and sustainability in agriculture.Besides the crucial importance and significant contribution of agriculture in overall GDP of agriculture-based developing economies, the situation of farmers especially the small and marginal ones is still vulnerable.Farmer distress and suicides are very common in countries particularly India and China.Both natural and manmade factors are responsible for such acts.The natural factors include loss of income due to natural calamities of flood, drought, crop failure due to prevalence of pests and diseases etc. which are not in control of mankind; whereas manmade factors can be controlled to an extent and include factors like burden of debt, low return for production due to inefficient marketing and unavailability of resources, higher cost of production due to use of outdated technologies in the production process.Besides considerable efforts and interference by the respective governments, the non-institutional sources of credit have deep roots in the rural credit markets of such economies thereby jeopardizing the prevalence and growth of institutional sources of credit.Several welfare schemes initiated by the governments like loan waiver schemes during crop failure, agricultural crop insurance etc. are not available to vulnerable groups including small and marginal farmers which actually need it thereby making their situation miserable.The exclusion of several vulnerable groups from the formal financial services may lead to their social exclusion in the long-run.The provision of timely access to adequate credit may not only help to uplift the situation and living standards of the farming communities but may also raise the production levels in an economy thereby accelerating its GDP growth and hence sustainable development in the long run.The question of whether it is a matter of choice or compulsion to use non-institutional sources of finance is still vague.In light of the above, this paper is an attempt to analyze and present various issues pertaining to agriculture credit at a single place.Relevant literature has been reviewed, classified and presented based on several themes.Literature on the following issues pertaining to agriculture credit has been presented and analyzed in this study -determinants of source and quantum of agriculture credit, gender issues, status, performance and current issues in agriculture credit, impact of agriculture credit on productivity, rationing in rural credit markets, repayment issues in agriculture credit, impact of reforms on agricultural credit and emergence of alternative sources of finance in agriculture.

Objectives of the Study
The broad objective of this paper is to present the studies on agriculture credit in an organized and easily interpretable way by systematically arranging various published and unpublished studies.The idea is to segregate the reviewed studies into suitable categories based on year, focus area, source of publication, country of study and analyze them accordingly.This paper can serve help for future researchers, policy makers by presenting several inter-related aspects related to agriculture credit at one place.Studies related to issues of determinants of agriculture credit, rationing in credit markets, repayment issues and certain related issues have been analyzed.In particular, this study may help policy makers/bankers/lending institutions of the country on which they are based in taking balanced review of status and performance of agriculture credit.A total of 110 studies published across the world on developing economies have been reviewed for the period 1995 to 2015.The paper also suggests the future prospects for research in agriculture credit markets of these de veloping nations.

Break-Up of Literature on Agriculture Credit
After carefully examining all 110 papers, the next step was to classify the literature into suitable categories with internal homogeneity within each category.A comprehensive snapshot of all the reviewed studies is given at end of the paper (Table A1).All analysis, tables and figures of this paper are based on Table A1.On the basis of review, the literature has been classified according to the following themes: (1) Year-wise classification of studies; (2) Region-wise classification of studies; (3) Country-wise classification of studies; (4) Source-wise classification of studies; (5) Classification based on type of research.

Year-Wise Classification of Studies
Table I gives the year wise classification of reviewed studies for the period 1995 to 2015. Figure 3 is a graphical presentation of year-wise publication of reviewed studies.A glance at this figure shows that more papers have been reviewed from recent years thereby indicating the increased importance gained by agriculture sector in more recent years.The maximum no. of reviewed studies is for the year 2012 (19), followed by the year 2013 (16) and 2014 (14) (figures in parenthesis show the no. of studies reviewed during that particular year).It can be noticed that the number of studies have been increasing particularly after the year 2006.In the Indian context, it was the period when its government attempted major efforts to promote financial inclusion in the country.

Region-Wise Classification of Studies
The reviewed studies have been further classified on the basis of geographic region of the world to which a particular country of study belongs.The maximum number of reviewed studies is from East Asia (6), followed by South Asia (4) and East Africa (4) (figures in parenthesis show the number of studies reviewed for the respective region).In East and South Asia regions, majority of the population is dependent on agriculture and therefore a lot of research has been done on agriculture sector of these regions.

Country-Wise Classification of Studies
We have further classified the literature on agricultural credit on the basis of country on which the study is based.
Table II shows the list of countries on which the reviewed studies are primarily based.A good number of researches have been carried out in various parts of the world.India tops the list with 36 studies, followed by Nigeria (20), Pakistan (09) and so on (figures in parenthesis represent the number of studies in that particular country).Figure 5 shows the country wise publications of the reviewed studies.It can be observed that agriculture sector in developing countries is gaining a reasonably good attention of researchers.

Sources of Publication
Table III gives a comprehensive snapshot of distribution of various sources of publication from where the studies have been retrieved.The most frequent sources of publication are Agricultural Economics Research Review (9); followed by Journal of Development Economics (6), Economic and Political Weekly (5) and so on (figures in parenthesis represent the number of publications in respective journal).A variety of journals have published research in agriculture as the agriculture sector is related to several other sectors of the economy and it affects and gets affected by changes in these sectors.

Classification Based on Type of Research
Here the reviewed literature has been classified on the basis of methodologies adopted in research.For this purpose, the literature has been classified into four categories namely-conceptual, descriptive, empirical and exploratory cross sectional studies.Conceptual studies are those which cover the basic and fundamental concepts of functioning (in rural markets).Descriptive studies give explanation and description of status, content/process and performance issues.Empirical studies cover data from existing sources to estimate and evaluate relationships among various variables.Studies based on primary data collected through surveys are defined as exploratory cross sectional.Table IV gives the distribution of studies according to type of research.Figure 6 shows the percentage distribution of the same.Most frequent methodology used in the literature is exploratory cross sectional (61), followed by empirical ( 22) (figure in parenthesis represent the number of studies employing a particular methodology).

Focus Area of Research
This section of paper presents the break-up of literature on the basis of the focus area of research on agriculture credit.Table V represents focus areas of the reviewed studies.Identifying determinants of agricultural credit is the focal point of most of the studies.Notations "a-i" show the bifurcation of focus area in literature."a" represents the determinants of sources and amount of agricultural credit, "b" denotes gender issues in agricultural credit, "c" depicts the status, and performance of agriculture credit in developing nations, "d" shows the impact of agriculture credit on output and productivity, "e" represents studies which focus on rationing of rural credit markets, "f" represents studies related to repayment issues in agricultural credit, "f" shows studies focusing on the role of Islamic Banking in agriculture credit, "g" depicts studies on agriculture credit via micro finance institutions, "h" represents studies which focus on the performance of agriculture credit in developing countries during pre and post reform periods.Figure 7 depicts the percentage distribution of focus area of reviewed studies.

Determinants of Agricultural Credit
Several researchers in the past have tried to identify the factors which significantly influence the household 's decision for choice of a particular source of agricultural credit (Akpan et al., 2013;Salami & Arawomo, 2013;Yuan et al., 2011).Several variables (factors) have been used in the literature by eminent researchers to analyze their impact on farmer household's decision.

Factors Affecting Quantum and Source of Agricultural Credit
On the basis of observation of reviewed studies, we found a variety of factors which have significant impact on a household's decision to opt for a particular source of credit.On the basis of review, these factors were classified into three categories based on common attributes as depicted in Table VI.The most common individual specific factors include caste, education, marital status of the household, contact with extension agents, years of experience in farming, land size, gender, contact with large landholders etc. (Aliero & Ibrahim, 2011;Dzadze et al., 2012;Sebopetji & Belete, 2009;Akudugu, 2012;Akpan et al., 2013).

Techniques Used to Identify Determinants of Agriculture Credit
Different researchers used various techniques to identify the determinants of access to different sources of agricultural credit.Table VII gives a brief view on techniques used in the literature.A large number of researchers used logistic regression to determine the impact of various socio-economic variables on the access to credit (Chauke et al., 2013;Ololade & Olagunju, 2013;Hananu et al., 2015).Logistic regression model can be classified into the following: (1) Binary logit model; (2) Multinomial logit model; (3) Ordered logit model.
While binary logit model is applied where the outcome has binary outcome (either 0 or 1), multinomial logit regression is applied where the outcome has more than two categories (Mpuga, 2010).Here the choice of reference category is arbitrary and this can be used as base category to facilitate comparison between "N" numbers of groups.While ordered logit model is used where the dependent variable has more than two outcomes (categories) having sequential order (Nouman et al., 2013).The outcome i.e. odds ratio gives the magnitude of change in dependent variable for changes in various independent variables (Kosgey, 2013;Baiyegunhi & Fraser, 2014).
Probit modeling has also been used by a fairly good number of studies to model the probabilities of access to a particular source of finance (Sen & Prajapati, 2013;Datta & Ghosh, 2013;Sebopetji & Belete, 2009).Both logit and probit models have been used in the literature.These are specific cases of modeling when the dependent variables cannot be measured on a metric scale rather it is categorical in nature (Bhanot et al., 2012).Pal and Laha (2014) used quantile regression along with Ordinary least squares method to get an estimate of total credit across various conditional quantiles of borrower groups.

Gender Issues in Agriculture Credit
Studies have investigated the impact of gender on the quantum and sources of agriculture credit.The formal credit was found to be biased against women.The factors affecting the choice of source of finance are different for males and females (Jeiyol et al., 2013;Akugudu et al., 2009).Goetz and Gupta (1996) assert that self-financed institutions in rural areas are more concerned about the quantitative aspects of granting credit to women, while the qualitative aspects like use of credit are not taken care of.Kabeer (2001) documents that loans directed to women have more chances of improving their personal and social benefits.Women are biased than men in terms of access to credit and such inequality is the most insidious form of inequalities (Schuler et al., 1996;Ogunlela & Mukhtar, 2009).

Impact of Agriculture Credit on Output and Productivity
Literature seems to be divided on the issue of the impact of agriculture credit on agricultural output.On reviewing the literature, it was concluded that the studies can be segregated into two categories: (1) Agriculture credit has positive and significant impact on agricultural output; and (2) Impact of agriculture credit on agricultural output cannot be directly established; While some studies in literature find the direct and significant impact of agriculture credit on output (Bashir et al., 2010;Iqbal et al., 2003;Saleem & jan, 2011;Rima, 2014;Villanueva, 2014;Ekwere & Edem, 2014), some others hold that the impact of agricultural credit on farm output cannot be directly established (Sriram, 2007;Hussain, 2012;Zuberi, 1989;Sjah et al., 2003).Ahmad (2011) and Raza and Siddiqui (2014) insist that it is indirect credit to agriculture which has significant impact on agricultural output and not direct credit.De rosary et al. ( 2014) used simultaneous equation modeling to see the impact of credit on economic functions like production, consumption and investment of households.Duy (2012) applied stochastic frontier analysis and quintile regression and found positive impact of institutional and non-institutional credit on farm output and production efficiency.Similarly, Xi and Li (2007) used quintile regression to see the impact of formal and informal credit on income and efficiency.Binam et al. (2004) estimated technical efficiency of various categories of farmers and found that efficiency differences are significantly influenced by the amount of agricultural credit utilized in production.Technical efficiency of farmers includes factors like flow of information, access to better infrastructure facilities, farmer's expertise in management of resources and availability of required funds (Iqbal et al., 2003;Chisasa & Makina,2013).Obilor (2013) applied regression analysis and found that credit allocation to agriculture had significant positive result on productivity.Dong et al. (2010) used probit modeling to determine the relationship between various socio-economic variables and credit condition of households and found that agricultural productivity can be improved with increased use of credit.Owuor and Shem (2012) used switching regression model which is estimated by employing Heckman sample correction method and found significant impact of agriculture credit on production and various input use.
A large number of studies have employed co integration to see the causality between agriculture credit and output.While some findings suggest positive significant impact of agricultural credit on output (Ammani,2012;Okulegu et al., 2014;Ihugba et al., 2013), some other studies reject this hypotheses (Oyakhilomen et al., 2012;Musuna & Muchapondwa, 2008).Table VIII shows the distribution of studies used in the literature to analyze the impact of agricultural credit on farm output.In the literature, Cobb-Douglas production function has been widely used, followed by Granger causality and Co integration.The other approaches include correlation, probit modeling and mixed approaches.
Researchers also employed stochastic frontier analysis to see the impact of agricultural credit on productivity (Liu, 2006;Dolisca & Jolly, 2008;Nisrane et al., 2011;Kebede, 2001;Chiona et al., 2014).Several studies have used Cobb-Douglas production function to see the impact of agricultural credit on productivity (Sriram, 2007;Bashir et al., 2010;Iqbal et al., 2003;Saleem & Jan, 2011, Rima, 2014).It is a production function which represents the relationship between output and a number of input variables (Chisasa & Makina, 2013).To see the impact of credit or other variables, it is log-transformed to take the following form: Here  represents the log of agricultural output, β0 represents constant term, β1 to βn are beta coefficients presenting partial elasticity of various explanatory variables, εt represent random error term.Figure 8 shows the most common input variables used to determine their impact on output.Sial et al. (2011) and Iqbal et al. (2003) used dummy variable along with others to see the impact of various uncertainties like drought or floods on agricultural output and found significant negative relation showing decrease in agricultural output during bad years.

Rationing in Credit Markets
Literature is full of evidences to show that well-functioning and efficient rural credit markets can promote rural household's income level and thereby reduce poverty by promoting equitable distribution of resources.Besides institutional setup of rural credit markets, a large number of non-institutional lenders are also present in rural credit markets.As credit from institutional sources is rationed in these countries, rural households are constrained by credit severely (Rui & Xi., 2010;Kochar, 1997;Carter, 1988).Credit rationing is a situation when those who need credit do not get it in adequate quantity (Jansson et al., 2013).A large number of studies have tried to determine the reasons why the credit markets are rationed (Hashi & Toci, 2010;Weber & Musshoff, 2012;Jaffee & Stiglitz, 1990;Petrick, 2005).
The question that "is it choice or necessity" to resort to non-institutional sector has been addressed by several authors differently.Chaudhuri and Gupta (1996) assert that market for informal loans is created due to delay in disbursing formal loans and that the effective interest on loan from formal sector is same due to incorporation of bribe amount in formal credit which is paid by farmers to avoid delay.Beaman et al. (2014) find that large landholders who have higher returns to their investment choose their source of finance independentl y since institutional and non-institutional lenders both are ready to lend money to them.Kochar (1997) finds that the extent of rationing is much lesser than what it has been assumed to be and that credit supplied to rural households is less because it is not much demanded.Basu (1997) asserts that credit by formal sources is rationed due to the inherent risk present in agriculture and allied activities, thereby reducing the probability of earnings.Further it is the "congruence of interest" between landlord and tenants which gives birth to loan agreement.Bose (1998) argues that when moneylenders are not fully aware of the likelihood of default by various classes of borrowers, the provision of subsidized credit by banks can lead to adverse "composition effects" which deteriorate the availability of loans in unorganized sector.Guirkinger (2008) found that it is not rationing by formal sector rather lower transaction costs which drive rural households to informal sector.Such lower costs are enjoyed by informal lenders due to proximity and economies of scope.
Credit rationing has deep roots in agriculture sector as compared to other sectors of an economy (Weber & Musshroff, 2012).Rationing of credit causes a significant loss in income levels and consumption expenditure of rural households (Li et al., 2013).Stiglitz and Weiss (1981) explain credit rationing in terms of agency issues: (1) adverse selection; (2) moral hazard.
Adverse selection occurs in rural credit markets when the formal credit institutions are not fully aware of borrower's credit worthiness and therefore credit worthy borrowers are left when banks try to mitigate the risk of default by raising the rate of interest (Klonner & Rai, 2005;Binswanger & Deininger, 1997;Ghosh et al., 2000).While moral hazard occurs as a result of dominance of large and wealthy landholders while obtaining cheap credit since they possess more resources to offer as collateral and as a result the poor borrowers are left away (Simtowe et al., 2006).Figure 9 has diagrammatically presented the agency problem.

Repayment Issues in Agriculture Credit
Apart from rationing in credit markets, studies have also tried to analyze repayment issues in agriculture credit.Banks do not lend to poor groups due to fear of non-repayment and increase in their non-performing assets.A large number of studies have tried to investigate various factors which affect repayment performance of borrowers in rural markets.Kohansal and Mansoori (2009) applied logit regression to identify facto rs influencing repayment and found that farming experience, income of borrower, loan size, value of collateral offered as security have significant positive impact on repayment performance of borrowers while interest rate, total application costs and number of installments to repay loan impact it negatively.Figure 10 shows the classification of factors affecting repayment schedules.Various factors affect repayment performance of borrowers in the rural credit markets.We have classified these factors into two broad categories, namely: (1) Social factors; (2) Economic factors and; (3) Contract-specific factors.
While the most common social factors affecting repayment rate among borrower households are age, education, gender, marital status, experience of the household, household size, diversion of loan due to family commitments, incidence of crop diseases and pests, farm size, monopoly power created by informal lenders in markets, use of modern machinery and equipments, contact with extension agents, social relations of the borrower households.
Economic factors include interest rate on loan, income of the household, loan size, value of the collateral offered as security, total application costs, off-farm income, net profit, market price fluctuations, market value of livestock, fluctuations in commodity prices, amount spent on hiring equipment (Kohansal & Mansoori, 2009;Weber et al., 2014).
Contract-specific factors include various terms and conditions specific to a particular loan contract like lender 's supervision on utilization of loan, number of repayment installments, down-payment of loan, length of waiting time for receiving the loaned amount from lender, length of repayment period.

Policy Issues and Implications
In line with the objective of presenting and classifying the reviewed studies, several issues relating to agriculture credit in developing countries have been discussed.The review presented in this study can be used by policy makers/banks/researchers to judge the performance of agriculture credit in these nations and analyze the situation of agriculture sector in this direction.A review of the relevant literature highlights that the interest of researchers has been growing towards this topic over the past few years.However, there exist huge disparities in the number of publications in these countries.Determining the factors which affect household's choice of a particular source of finance has been heavily emphasized by eminent researchers in the literature.Several factors have been used to study this relationship at the micro-level, the most prominent of which are-literacy, land size, marital status, distance from a lending institution, age of the borrower, caste, religion, the value of assets owned by the household.A majority of studies have reported significant impact of literacy, size of landholdings and household assets on opting for the source of financing agriculture.Their findings reassert the importance of literacy especially financial literacy in covering the hitherto deprived groups under the ambit of institutional setup of an economy and in uplifting their living standards.Household assets and size of landholdings represent the value which the households possess and can offer as collateral security while procuring loan.As b anks perceive clients with more asset value as credit-worthy, they are more inclined towards such borrowers than their other counterparts.Therefore, it is important for the governments and other regulatory authorities to keep a regular and timely check on lending activities of financial institutions covered under their ambit and to encourage banks for social banking initiatives rather than class banking.In the development process of an economy, it is important to implement policies at the bottom level rather than at the top only.Next, the impact of direct institutional credit is found to be associated with the productivity levels by majority of studies which pinpoints the discrepancies in indirect credit mechanism.The failure of cooperative banking in India is one such example.Banking institutions in an economy should be promoted in such a way that they are sustainable in the long-run and their dependence on donors/state governments is minimal.Micro-finance and Islamic Banking appear to be alternate source of financing agriculture and find the mid-way of institutional and non-institutional setup of rural market markets.But micro-finance institutions are perhaps in their nascent stage and their management needs to be nurtured.Lack of awareness among masses about their working and lack of trust pose restrictions on their financial and operational sustainability in the long-run.

Conclusions and Way to Future Research
This study has reviewed 110 research papers on agriculture credit between 1995 and 2015 from various journals, working papers and several other published and unpublished sources.Studies on developing countries relating to agriculture credit have been systematically presented and reviewed.Agriculture credit is a topic of considerable interest in countries particularly India, Nigeria and Pakistan.After reviewing the studies, a noticeable growth in the number of studies have been observed particularly after the year 2006.The agriculture sector has assumed more importance in recent years amid rising concerns about food security and population pressures.In the Indian context, this was the period when government focused extensively on the increased use of institutional credit and several measures policy measures were initiated to promote financial inclusion.Majority of the studies are focused on determinants of sources and amount of agriculture credit by employing exploratory cross -sectional research.However, less research has been done on identifying the supply side determinants/constraints of agriculture credit.This could be an area of future research.So far, the implementation of policies framed by the government has lacked the desirable commitment from banks/financial institutions which are in close proximity with farmers.It appears that in order to make the credit delivery system inclusive, efforts need to be initiated at the bottom level rather at top.Further, this study is limited to developing countries only, therefore future research can be undertaken by including developed nations which remain uncovered in this study.Studies by Duy (2012), Liu (2006) applied stochastic frontier analysis to analyze the impact of agriculture credit on farm output and production efficiency.Application of stochastic frontier analysis in agriculture output is paving a new way for future research.To analyze the level of integration between credit and output, new techniques like Auto regressive distributed lags (ARDL) can be applied in future research.Micro-finance institutions appear to be a good option which can fulfill the gap for institutional credit but micro-finance institutions themselves are faced with internal conflicts of interest and discrimination and paucity of funds at their end.Literature seems less focused on reporting such discrepancies and how MFIs can be made sustainable.The policy implications and impact of microfinance on agricultural production and household income can be studied further.

Table 1 .
Year-wise classification of reviewed studies Source: Based on author's own calculation of reviewed studies.

Table 2 .
Country-wise classification of studies Others include Indonesia, Ethiopia, Uganda, Iran, Malawi, Thailand, Algeria, Kenya, Peru, Mali, Mexico, Malaysia, Madagascar, Philippines, Cameroon, Yemen, Barbados and Nepal.Source: Based on author's calculation of reviewed studies given in TableA.

Table 4 .
Distribution of studies based on type of research

Table 5 .
Focus area of literature Source: Based on author's own calculation based on studies reviewed.

Table 6 .
Classification based on determinants of agriculture credit

Table 7 .
Classification of studies according to techniques used Others include Tobit analysis, discriminant analysis, general linear model, ratio analysis, mixed approaches.Source: Based on author's own calculation of reviewed studies.

Table 8 .
Distribution of studies based on methodology used to study impact of credit on output Figure 8. Classification of various input variables used in various production functions Note 1. * means credit used for various inputs like seed, implementation of machinery and tube well, fertilizers, pesticides & fungicides, land preparation other costs.level,loanamount, net income from farm, farm size, sideCross sectional job, effective supervision have positive impact on repayment while gender, marital status and household size impact it negatively 93Weber et al. (2014) (Madagaskar)