Report No. 436 Bank Policy on Aorici iltu irn Credit May 1, 1974 An.,icu!ura a2nd Rua!m rvlopinnmant Drnirtment Not for Public Use Do-umet of tka Intpenmt6nnAl Rank fioe RD nctnirtion anfi 1eve!ninnmant This report was prepared for official use only by the Bank Group. It nMay not be published. auoted or cited without Bank Group authorization. The Bank Group does not'accept responsibility for the accuracy or completeness of the report. BANK POLICY ON AGRICULTURAL CREDIT Tabule of. uContents Page No. PREFACE . ...................................... iv-v CONCLUSIONS AND RECOPOIENDATIOUS .... .......... 1 Dimensions of Credit Problems ....... * ... 1 Size and Nature of Bank Group Operations 2 Impact on Output and Incomes ........... . 2 Credit Policies ................. 3 Eligibility and Security .............. 4 Interest Rates .......................... 5 Repayment Performance .............. 9 Credit Channels ............... 10 Recommendations ........... .. ............ 14 AGRICULTURAL CREDIT PRACTICE AND PROBLEMS .... 17 1. Evolution of Current Approach .... ....... 17 Volume and Nature of Insti utional Credit 18 Development of Bank Group Participation 19 Categories of Bank Group Lending ...... 21 2. Problems of Agricultural Credit Markets.. 22 Limitations of Informal Sources ....... 22 Changing Credit Uses and Requirements.. 25 Mobilization of Savings ............... 26 3. Constraints on the Role of Credit in Development ....... ...................... 26 Limited Applicability of New Technology 26 Economic and Social Constraints ....... 27 Non-Credit Policy Requirements ........ 29 AGRICULTURAL CREDIT PROGRAMS AND POLICIES .... 31 1. Designing of Credit Programs ........... 32 Separation of Programs for Different Borrowers ....... - ... o- .......... 32 Providing Credit for Particular Purposes 32 Terms and Conditions of Lending .. ..... 33 Provision of Complementary Services ... 34 Table of Contents (Cont'd) Pa,e No. 2. Financial Viability of Credit Institutions 36 Loan Delinquencies - Overdues and Defaults .............................. 36 Costs of Agricultural Credit ............ 39 Interest Rates ... ....................... 40 ITT. SYSTEMS FOR DELIVERING AGRICULTURAL CREDIT ...,. 46 1- po7nlt-v-Makina and Adminintrative Issues . 46 2 rhAnnnl- of Agritul t-iirnl Finanre =.==.==== 48 3. Agricultural Gredit Agencies =49 Aoricultural. .and .lnplunlpmpnt RBanks 49 Farmer Groups.and Cooperatives .... ...... 49 Project AuAthnrities.51 .Non-Institutionalized Commercial. Channels 52 .Institutional -CG ercial:.Cha.n-els . ._ 5 -4. r-uidelines for .Delivery Systems of Agricultural .Credit .54 Larger Farmer Channels ................... 54 Small va.mer CS-hanells ................................. 5 ANNEX TABLES 1. Institutional Lending for Agriculture 2. Distribution of Agricultural Loans by Type of Lender 3. Farmers Receiving Credit from Institutional Sources 4. Bank Group Agricultural Credlit and Total Bank Group Commitments for Agriculture. FY 1948-1973 5. Bank Groun Agricultural Credit Onerations; by Funding Aeencv. FY 1948-1973 6. On-Lending to Farmers and Number of Beneficiaries in Bank Group Agriiu1rurq1 CGredit Onpratinns by ';ize of Farm- FY 1969-1973 7. Bank GrnOin Arrirtl ttiral Credit Operatinns bv Country and Per Capita GNP Level, FY 1948-1973 8. Contribution to Project Costs by Bank Group Agricultural Credit Operationns hb Per ranita GNP nnt-rv Gateonriei, FY 1948-1973 9. Interest Ra-tes to Farmera hv Soiurce nf Lnana 10. Bank Group Agricultural Credit Operations, by Major End HUo by Region, FY 1948-1973 11. Duration of Loans Made by Selected Institutions 12. Measures of Loan Delinquency of Selected Institutions 13. Administrative Costs for Selected Institutions 14. Bank Group Agricultural Credit Operations, by Lending Channel to Ultimate Broe,F 9817 - . I l. BANK POLICY ON AGRICULTURAL CREDIT PR -IACE Bank lending for agricultural credit has expanded considerably in recent years, ar.dul credlt is now- th.e largest component of JBanLI agricultural lending. In the course of this expansion, a number of issues have arisen adu problems LhLave bUeen encountered, andU these are likely to bUe compounded as increasing emphasis is placed on lending to small farmers. Accordingly, this paper exsamines: (i) thLLe issues a.d proLems associatedU with the use of credit on the farm and with institutional lending and (ii) the policies _t __ _2 _L_ 1_ _=I_ t.. - 1t.T__'1 .3 D__1_ a__ --I _1_ ~J_ _t ___ - _ _L - II tilat Mighlt bDe considered uy Lth rlUL.XU D4Ban GroUUP dlLiLngLS LLthose WllhLCLI W1ill facilitate the provision of credit to large numbers of small farmers. Small farmers, as considered here, include families farming less Lhan five hectares or, fln those countries where all farms are small in absolute size, those farmers who comprise the lower half of the country's cultivators. Since these constitute tne core of the poverty problem in the developing world, improving their productivity and incomes is a matter of high prio-ity. But credit has also to ue provided to medium- aduU large-sized farmers in order to help meet the world food problem. The evidence suggests that holdings of 5-50 nectares are also short of accessibie production credit. Bank credit projects in the agricultural sector provide more than finance for on-lending to farmers. Funds are provided for processing and marketing farm produce, forestry development and fishing enterprises. Pro- vision is also made for the support of technical services and training, feasibility studies, project preparation and other institutional assistance. While these are significant elemencs in Bank lending the main problems- and the focus here - concern lending for on-farm development. Data used in this paper were drawn from Bank reports as well as from the many studies that have been made of farm credit. Both sources of information leave much to be desired. The data from studies are generally incomplete and often of poor quality. In particular, little is known of the extent of farmers' needs for institutional credit, of the uses to which borrowed funds are put, or perhaps most significant, of the impact of credit on output and productivity. It is argued that non-institutional sources of funds lend more than the formal institutions in many countries; but there is little information about lending terms and conditions in the informal market, and what is known is derived from observations by knowledgeable people rather than from hard evidence. Most of the Bank data come from appraisal reports. Since these are prepared prior to project implementation, they present expectations which may or may not eventuate. For example, these reports show the average size of expected beneficiaries for each project but there is as yet little basis for comparing achievements with those expectations. Supervision reports - iv - yield little in most cases because information is generally not available on what happens at the farm level in terms of the actual size distribution of borrowers, the uses made of the funds and their effects, or the repayment record by different classes of borrowers. The information the Bank has on such key institutional issues as the costs of administration and on defaults is both iricomnlete and not comnarable between nroiects. This situation will be rectified to some extent by the review of five agricultural credit projects heina undertaken by the Onerations Evaluation Denartment. Despnite these aerinus shortcomings in data- the statistics nraeented are believed to reflect reliably the overall magnitudes on size of programs, extent of coverage and intrt rAtes charged. The informatinn nf institu- tional arrears and defaults and the costs of administration contain a greater marin nof avvnr. Tnformatinn nn the imna-pa of ovredit- no untt, ane nn the sources and charges for non-institutional credit is even less reliable. Thus, this brief preface highlights the need fnr hbttev inf4nvmn-tin onr nll nanpcht-s of rural credit and its impact on production and incomes. ttt -TIa Tf'7 ,XTh Af'nfTTI1T PT'TTflAT rr'n"T% T "I DIUU4N rulL%L1I VA4f IU~~~l.VL1± CUiLCLSIUO AND KRECOurLU'AJTITNS 1. Credit is often a key element in the modernization of agriculture. Not only can it remove a financial constraint bUL iL may providCe tLIC ±UC1- tive to adopt new technologies that would otherwise be more slowly accepted. Credit facilities are also integral to the process of commercialization of the rural economy. However, no amount of credit, even at the most reasonable rates, can provide a panacea for raising farner productivity or improving incomes of the rural poor. Success in this aepends on many factors, including the availability of complementary inputs and services, sound credit policies, well-managed institutions and appropriate delivery channels. These issues and their implications for Bank operations are discussed in this paper. 2. Although credit problems have been the topic ot much keen deDate and a voluminous literature, hard information on the structure of credit, in terms of sour,es, recipients, rates, uses, mechanisms and impact, is frag- mentary and leaves much to be desired. The picture collated in this paper from Bank and many other sources is necessarily rough and presents only sketchy outlines of some essential features. Dimensions of Credit Problems 3. Outstanding institutional loans to agriculture in the LDCs are estimated at about $15 billion. Total agriculture credit outstanding in these countries is unknown. The bulk of it originates in the informal sector and it is probably not less than five times the estimated outstanding institutional credit. 4. The percentage of farmers receiving institutional credit varies widely in different parts of the developing world. In certain African countries around one percent of the total number of farmers use institutional credit, while in Taiwan nearly all farmers have access to it. About five percent of farmers in Africa get institutional credit while coverage in Latin America and Asia (excluding Taiwan) is about 15 percent of all farmers. Large farmers have been the main beneficiaries of institutional credit. It is common to find that 70-80 percent of small farmers in a given country have virtually no access to institutional credit. 5. Most of the credit that is available to farmers in developing countries is for short-term loans for a season or for up to one to two years. These credits are used for purchasing current inputs such as seed, fertilizer, pesticides and the like. There are some credits available for 2-5 years for purchasing livestock and some longer-term credits are available for acquiring items such as tractors or pumps for irrigation. The available supply of credit to all farmers, though, is heavily skewed in favor of short- term credit and this skewness is even more pronounced in the distribution of whatever credit is available for small farmers. Even though most of the limited credit available for small farmers is short-term, the supply is - 2 - inadequate especially as it i-s creAdit of this kir.d thtismstueddb small farmers to produce a marketable surplus, the first requirement if they are to contribDutete to e development process. ize. adu Nature of Bank Group 0U p-4 tiUn 6. Total Bank Group commitments to agriculture rose from $468 million in the 16-year period FY48-63 to $621 million during FY64-68 and to $2,589 million in the most recent five-year period. The credit component rose from 20 percent of total lending for agriculture in the first period to 56 percent in the last five years. There has also been a significant shift in the snare of credit funds going to countries with lower per capita incomes. In the most recent period, agricultural lending Lo tne poorest countries exceeaea $1 billion, of which more than half was for farm credit. Approximately one quarter of all credit that has been financed by the Bank/IDA is intended for small scale producers. 7. The Bank Group's initial objectives, namely to increase agricultural production through the economic use of resources and to develop agricultural credit institutions, resulted in a concentration at the outset on commer- cially viable farms and related enterprises. Recently, however, there has been a significant shift in emphasis in the allocation of Bank Group resources to small farmers. Project appraisals during Fi69-73 envisaged ,hat long and short-term agricultural credit operations would benefit over 900,000 small farmers with holdings of five hectares or less located mainly in India (long term credit for 300,000 small farmers) and in Ethiopia (seasonal credit for 400,000 small farmers). While the numbers have grown in recent years, Bank Group operations have probably reached only one percent of the 100 million small farmers in the developing world. 8. The Latin America region has been the main area for Bank Group farmer credit operations with $507 million in FY69-73, followed by Asia ($442 million), EMENA ($258 million), East Africa ($140 million) and West Africa ($61 million). Two countries, India and Mexico, together received nearly two-fifths of all farm credit lending during FY69-73. Impact on Output and Incomes 9. Available evidence demonstrates that credit stands little chance of being ursed for productive purposes unless accompanied by certain other. conditions. These include: (i) clear opportunities for economic gain from adoption of new production technology or other improvements; (ii) widespread recognition and acceptance of such opportunities on the part of the farmer, along with access to training in the necessary skills; and (iii) delivery systems which provide ready and timely availability of the inputs required, and market outlets for farm production. - 3 - 10. These conditions do not often prevail in LDCs. To be successful in expanding production, each of the constraints, whether financial or non-financial. must be relaxed. This may involve new varieties and breeds. improved water control, basic infrastructure such as access roads, reliable provision of inputs, or an effective marketing infrastructure for farm produce. Price policies, such as unrealistic exchange rates, export duties or artificially low prices designed to favor urban consumers, can also reduce the profitability of an increase in output. If farmers are ignorant of the economic onnortunities onen to them or how to take advantage of them. the lack of extension services may be the limiting factor. Alternatively, if the constraint is rink aversion, sunnort nrices and nossiblv cron insurance could contribute to the adoption of a new technology. Especially for small farmers, it is esspntial to nrovidd a romnrehennive nackage if the notential for increased productivity is to be translated into a commercial reality. 11. Different types of potential innovations may each have different finnnri:l rpnitirp.npntc rponrdlp Q nf thi Si7.p nf hnldinac nf hnrrnrwpr Snme innovations, such as a switch from one variety of seed to another, may requniiire lit-tlo atditionnal rnnital. Othpr rhAnaoa aiitrh na the intrrngutiorinn of irrigation pumps can be far more demanding in their financial requirements, an.d m.ay require more resources than tradit4ion1al source var nble to provide. But farm credit is usually a necessary if not sufficient condition for inroosita - aln - r^t%rti4%rSt-, nnA 4mnr%niroF inrntun T'hil a i oQ^natsinllv increased agr~cu rar pr---------- t a iproued ncomes. -Ths _ e cial - y---- so for small farmers who do not have savings or ready access to institutional s ou rces. There appears to be conrsiderable scope for the u,se, of credit to replace or augment credit from traditional sources in order to allevi-- -n---l s- i4tuationrs whi-h force interest rates to --fessi-mel- high levels; to overcome inelasticities in the supply of credit which become apparent w.her. rew opportn itiesh emerge;tas to eann the Safonfl financial problems of rural households; and, most importantly, to encourage small su b sistssence f a.aers to rais e teir output n .d enter the commercial sect-or. a ua, ..c,L aL .~a U .aa *& LLA4~ CUR. ..LL t.J M~'~ OJ tL Furthermore, land reform, if pursued widely, could sharply increase credit needs oil former tenants -which -were previously- supplied- by anlod6 J. &L.L LL ~~~~L V lUu.y aUI,j.L.L=U WY1U7OL S 12. 'ThI..,- e '--est ways to meet this Ad...and a-- st-1l the subject of ---peri I& 4. &I U t. L. L .~~ LLA& L~iIO L CLA= AL.LJ.J. LI ~~ j & ^ ment. There is little experience on which to base firm views concerning the ost ef L ectiLve credit policiLes ar. d appropriate len.ding channels ad i.stitu tions. There is a need, therefore, for trial and comparison of alternative approaches, arnd for further mUoni.torin.g and acquisition of data so as to pemUit flexibility in implementation and management of Bank projects. Credit Policies 13. Credit policy discussions center around four main issues: (i) eligibility criteria and security requirements; (ii) the level of interest rates and the merits of interest subsidies; (iii) repayment performance; (iv) credit channels. Eligibility and Security 14. Traditionally credit agencies hiave required that borrowers both large and small pledge some collateral, usually land, as loan security. This practice, and the low valuation frequently placed on lands, excludes tenants and militates against small farmers who often lack certified titles to their land. Obtaining and processing re'ated documents also substantially increases the cost of lending, delays the disbursement of loans and discourages lending to small farmers and borrowing by small farmers. Lenders are also discouraged because foreclosure is extremely difficult to implement and often politically unacceptable. 15. The Bank has consistently emphasized that the renavment capacitv of a borrower should be determined by the appraisal of the productive capacity of his holding and this should substitute for securitv as the essential criterion in loan decisions. For long-ternm credit and large farmers, insistenne on land as collateral is quite in order. For those who have the resources to repay but do not, there is nio real substitute to court action, at leact onn selected hasis ns exmnles to othersc For short- and mediim- term credit, chattel mortgages and liens or mortgages on crop production are nnnronriatp- they cost 1ess to dorume- t and ean he arronnmliished more readily than land mortgages. They are not yet more widely used because legal pro- -ediires are qtill tnn riTmhersnmp and the se,r-rity then ffpr is g enera1 1 v considered to be poor compared with land mortgages. One way to make crop liens more acceptable as an alternativre to use of land as collateral is to coordinate repayment with marketing of crops. This system is likely to be of pavrticula,r imn,rt,n,e _n Aealing wi th the sutycni-n eblamc of Qsmall fa,rm- ers. It is being used successfully, particularly with those crops which are subject to-n a mr.nnly ar.dR xli,4ch are centrally7 procensrse,A a tobacco, cottnn cocoa, tea, coffee and sugar cane. A variationi would employ contracts whereby t he r.edA t -f ennc,y i -- _.;,l darpe r .. c r e rt- f ih-, --F . -li Mer .S u t plu, * r at Lh *r, t- fixed amount. Another approach is to coui-l'.e crG,c insu-rance with credit in order to protect both the borrower anr. the c-:e-6:- ageny agair.st th v-u a of nature. Some success has also been achieved. by lending to groups (informal cooperatives) with thL e m.ebers held1 separctel ant oeveralty resp-onsbl for4 the loan. Solutions to the problem of collateral are important if smaller an teAn fimr ar to Dnft fr_ crui proA AraAiA ; ; tA1 >_ A ... kAA o in the future will lie in some fcrm of group responsibility for individual UUL1Uwiugs-ajl approach yet to ue devtpUUUx on a Large scale. WLatever solu- tions are developed it is important that every effort is made to minimize cumbersome proceaures for lending to small far.lers, 16. Small farmers and tenants are usually penalized by the cumbersome and time consuming procedures involved in app'lying for their loans. Many lending agencies have rigid procedures for processing loans, whether large or small. These include the completion of complex forms and a pre-audit of the borrower who, if he is a small, farmer, is often illiterate. Thereafter, before the loan is issued, an official has to visit the farmers' holding and when the loan is eventually made it is available at the lending institu- tion (which may be distant from the borrowers' holding). The repayment terms of the loan will often be rigid without the flexibility needed to accommodate the natural hazards of farming. There is great scope for innova- tion and modification of the rules and regulations that govern the require- ments for collateral and the procedures involved for borrowing by small farmers. The need is for simplification and flexibility to facilitate ready access for farmers to credit. This may only be possible with the modification of laws governing the granting of credit and with greater reliance on self management and policing of individual sub-loans by groups of farmers who assume responsibility for all the members of the group. 17. Farmers should generally be required to contribute to the costs of the investment for which they are borrowing. This emphasizes their respon- sibility to make it successful, increases the lender's security and spreads institutional funds further. While significant down-payments should be required from those borrowers who can afford it, great flexibility with regard to small farmers, such as acceptance of contribution in labor, should continue to prevail. Strict repayment requirements compatible with cash flows are also justified because of their role in resource conservation and generation. Interest Rates 18. Capital/credit markets in developing countries are imperfect in varying degrees and as a consequence interest rates may not allocate resources among competing uses as effectively as they should. Also, other price effects or exchange rate policies may swamp the impact of changes in interest rates. Hluzhi more needs to be known about the effect of differential interest rates on resource allocation between sectors and, within agriculture, at the farm level. 19. It could be argued though, in strictly economic terms, that interest rates in agriculture and elsewhere should cover the cost of capital and the associated costs of services to provide capital to the borrower. In that event the interest rate would represent cost of making capital avail- able so facilitating the allocation of capital in line with its most effective use. An interest rate that would cover the tost of capital would include: (a) The opportunity cost of capital: this represents the foregone opportunity costs of using funds for agricultural credit rather than for alternative programs. Estimates of opportunity costs for capital are seldom less than 8 per- cent in real terms, approximately the level required to mobilize savings effectively; (b) The costs of administering the provision of credit: these are costs that are directly attributable to the processing, delivering and administration of loans. An efficient insti- tution making medium and long term loans to large farmers can operate at an administrative cost of around 3 percent of its tntal nnrtfnlin (The median, administrativp cost for the lending institutions discussed in Annex 13 is 5 percent) -6- Costs of administration rise as the size of loans fall. as the duration of loans shorten and as accounting services have to be exDanded to cope with liree numbers of sma1 scale borrowers. An efficient small farmer credit insti- tiltion can onerate with administrative consr of hetween 7 percent and 10 percent of its total portfolio. These costs would bh for nrovidinQ a miy of short term and Inng term loans and would be for services associated with pro- cepsina And dipliuerina 1nAnq, cost-s wtoild not incuiide then provision of extension and other public advisory services nor-m.all madp availablh qas A nonn-niofi-( npuhlic se4lce (c) The costs of rickc anti defaiults-: costs of lending ha to incorporate an element to cover losses through defaults. The 14imited evidence avai-l-aibla 4idA4ates that, over time, the level of defaults is no greater among small farmers than among larger farmers. The evider.ce indicates that the more careful the scrutinizing of loans, the supervision of loans and the pursu4to de inu th floe the default rate (but the higher the administrative costs). Limited evidence indicates that- 'normal1 defaults can be expected to add 4 percent to the costs of lending. 20. In summary, total costs, in real terms, for an efficient institution could ble etween 15 percent and 22 percen dependirg on the nature of the oper- ation and the size of the loan. Costs for lending to small farmers would be at *-lie uppers enlu ovE h rang ans; A^s mayx well IE uc1 1silers. ~L LL~UjJJ~ ~LU IJL LLk LO15 ~iLL UA ~L J. U~tLLLA. I 21. These rates cor.tras; withL interest rates cor 'instLtutUonal credit to agriculture as a whole, which are typically low. Although there is consid- erable variation among countries anud institutions, most nomiual interest rates for agriculture range from 6 to 30 percent, averaging about 10 percent. After adjustment for price changes these translate into real interest rates of be- tween -16 and +16, with a mean of about 3 percent. Frequently, the real rate for agriculture, even if positive, barely covers administrative costs, iet alone makes lending profitable after allowing for inevitable defaults, and seidom permits paying an interest rate on deposits that will attract savings. 22. The available evidence, as given in Table 9 in the Annex, indicates that commercial lenders charge much higher interest rates than institutional lenders. Petty traders, money lenders, landlords, and some commercial bankers charge anywhere from 3 to 20 times as much as institutional lenders. Payment of suclh highi interest rates would seem to indicate that borrowers would be prepare(1 to pay an institutional interest rate of 15 percent to 22 percent-- a rate much lower than the commercial rate but still high enough to reflect the real cost of lending to small farmers. 23. There are, however, a number of major factors that require consid- eration before advocating raising interest rates for lending in agriculture in general and to small farmers in particular. These include: - 7 - (a) Effective interest rate policy has to take account of comparative intersectoral rates. In practice real rates throughout most economies are likely to be well below the real economic costs of lending. Forcing up interest rates for agriculture alone can lead to uneconomic diversion of resources and, not infrequently, may be negated by sub- sequent leakage between sectors. Correcting the problem for agriculture, therefore, rests on restructuring lending conditions as a whole; (b) Lower interest rates than those representing real costs may be used as a compensation or offset to adverse terms of trade between agriculture and industry or agricult'ire and the rest of the economy. There is some merit in this but the use of interest rates for this purpose always car- ries the danger of proliferating distortions that already exist in an economy. A preferred approach (where possible) would be to eliminate the discrimination or to find other (and better) avenues to offset the discrimination, e.g. through changing price policy or modifying those trade or fiscal Dolicies that encourage high cost industrv while penalizing agricultural exports; (c) Introducing high interest rates for agriculture may have a deleterious impact on the rate of arceptance of credit= Where agricultural credit rates are very low to begin with-- as is often the rasP--1arge and ahrbnt increases in the interest rates may have adverse psychological effect on far.mierQ' readineQQ tn uitilize tcred1it- The rate of rhange in interest rates has to be given careful consideration P.qnecinl1v for itr imnart on small neale farmern mnving into the money economy. Hence, any policy that contemplates raising interest rates may reniri-e thnt the change beimple- mented over a long period of time. 24. A further argument that is mooted against raising interest rates iC th,t thn A_ ,-rn nF o0,4.-.1t:1 - 4a IQ hN,a4 -n *hi "eeds of small farmers that it should be subsidized. This argument rests largely on social grounds - subsidized interest rates are vi Aed as an. irnstrument to tra.nsfer income to the lowest income groups and so help redistribute income. In view of the widespread poverty of the s.all-scale farmer-s ther- 4i undoubte mert- in transferring income to the rural poor and it may well be that the interest rate is a con-ven-iaent t.manhisa4.. r this purposea Aga4ir.s thU4ia rment (i) Subsidized creAdi canrAt ca ator the eofr re cesac ar.d can lead to excessive capitalization of farm investment includ- ing theLuse of lour dispac.g macir.ery. F h o if credit is used productively then there is evidence that far,mers--e-ven sruall 'arm-ers--car. 'ear the real co-s of crdLUt;L U IM L=4J. -UV0 U credit; - 8 - (ii) Subsidized credit for small farmers tends to result in leakages. Experience indicates that larger farmers tend to gain the greater proportion of whatever subsidized credit is available for use in agriculture: (iii) The provision of subsidized credit to small farmers is also open to corruption and political abuse. Subsidized credit intended for small farmers has often been used for financing non-agricultural ventures, frequently ven- tures in which local politicians use their influe:nce to acquire the low-cost credit. Political pressure is also often used to avoid renavments and to write off debts: (iv) Subsidized or low cost credit means that the rate charged by the lender does not cover the costs of lending so that t-he lending institution incurs a 'loss' on each loan; con- sequently, even with a high repayment rate there can be a reduction in the inflow of funds into Institutional crediit and a depletion of the resources of the lending institu- trion. The finanrial viability of the lending institution could be weakened and the level of lending could be reduced unless the 1noses incu,rred frnmn 1w interest- rat- were subsidized from other sources such as the national hbudgpt- Tt should be borne 4in m4i that income dis- tribution is not really improved by subsidizing further Hhnaamanl 1 fnvma lrrlvnAi, 4t tt±ng rv.n s4_ tutional credit when this reduces the resources available to the lending institution to provide similar access to other small farmers. Additionally, interest subsidies would not help the often large arnd een poorer e roup. u landless labourers; (v) Subsidized interest rates have also been advocated to help make sritLalL far.,.ers ruore productive. Y.- s i tend to be most effective when they are linked to intro- ducing a particular technological change. ThL.us, where it is deemed desirable to use subsidies to encourage change, it is prelerable to subsidize particular inputs related to that change rather than a general parameter such as the cost of credit. Subsidizing part'cular inputs such as fertilizer has the advantage that the cost of the input to the farmer can be varied over time, and the variation can depend on the acceptability of that input; subsidized inter- est rates, on t'ne other nand, have a pervas've efLfect ,LLat spreads beyond any particular technological change; -9- (vi) One subsidy that might be effective in avoiding some of the problems enumerated above would be a subsidy to cover the administrative costs of lending to small farmers. Such a subsidy could equalize interest rates for all loans of equal duration and so help foster optimal resource alloca- tion; at the same time unified interest rates would minimize leakages and opportunities for corruption. 25. The preceding discussion provides ample evidence that there is no simple or unique answer to the question of what is an appropriate interest rate for agriculture, especially for small scale farmers. Many considerations have to be brought to bear. In so far as the Bank is concerned, the Bank should be) working towards a long run objective of positive interest rates reflecting costs of lending; an intermediate objective might be to cover at least the opportunity cost of capital. Subsidies should, in general, be confined to cases where they can be clearly justified and are likely to be effective in view of the pattern of financing in the project area. 26. There are obvious difficulties in any single institution--such as the Bank--trying to obtain interest rates on particular projects that are different from those charged by the same or competing borrowing institution on other but similar projects. This difficulty is compounded by the fact that many external lenders have accepted the principle that borrowing insti- tutions should on-lend at subsidized interest rates. In these circumstances the role of the Bank would seem to be one of persuading the external lending agencies and the governments to follow an interest rate policy consistent with the needs of the agricultural economy. 27. It has already been stressed that credit is only one element in the package of inputs and services needed to raise the productivity of small farm- ers. In many instances credit in itself may not be the most significant element. The attitude of the Bank towards the question of an appropriate interest rate will therefore tend to be viewed in the perspective of the overall proiect; the attitude towards the interest rate will have to be influenced by the extent to which the prolect overall would achieve the obiective of raising the produc- tivity of large numbers of small farmers and giving a satisfactory economic return. Repayment Performance 28. Repayment performance by large and small borrowers on farm loans is tynicallv noor. However. while many large arrears are renorted. in the long run most loans are eventually paid and the default rate is much lower than the arrear rate. The mai0r problems for credit institutions associated with a chronic poor repayment performance by all farmers are loss of liquid- itv and the additinnal mannnper and Cnote invnlved in rnolletinn activities. The reasons for the large arrears are independent of size of borrower; they inrltde the vaanries nf nrnduetinn such as nriee and weather effen-ts socio- logical factors and most important poor organization and management of credit institutiions. Given the likelyl fluct-uaint-inns in farm incnme from year-to-year there is a strong case for flexibility in scheduling repayments. However, - 10 - thi-s should n.ot encourage t-arA4ress or provide less pressure on fam-ers to LL*..LO ~Li LU I IU- - tL U&_~fl LL M~O L make repayments when they are able. This requires a firm stance on the part IL'A in stLi tutions, w h4 Ch has seld4om been forthcomir.g, often as a result of political pressure. Closer supervision provides the most obvious means of minimizing defaults, and t to be excellent scope a higher supervision cost against a lower delinquency rate in many insti- tutions. * .uptie V-i on has the` aUdvantaage t1AtL it er.courage,s favorable ehLav- iour patterns which may eventually allow supervisory activities to be reduced wh'ereas unchecked del i qu enC -L= - or.ly snorg a aos l s tuat'on. "In Lthe W1L~~d~ ULII.11 .~~~eU ~ ~ ULLL. =LLL.VUUi42r, C WVLZ5rU±AL1r . LLU& L UL. ULJ case of small farmers, though, the costs of supervision will be high unless this can be done on a group basis with some collective responsibility for debt collection. Credit Channels 29. In considering appropriate and effective credit delivery systems we must come to grips with two major issues: (i) whether institutions should be specialised in providing credit alone, or alternatively be multi-purpose organizations providing marketing services, input supplies and other items and service along with credit; (ii) how to cope with large numbers of small farmers. Whether farm cre&it institutions should be specialized or multi-purpose is something of a moot point, but experience tends to favor multi-purpose ones. Both kinds have had their successes and their difficulties. Multi-purpose cooperatives have found it difficult to exercise credit discipline among their members in some cases, and some observers of the experience with cooperatives argue that the administration of credit should be kept separate. On the other jiand, specialized institutions have a tendency to remain aloof from the everyday problems that affect the financial needs and repayments capability of their clients. Since credit is often only a component in a package, multi-purpose institutional arrangements are generally to be preferred at the primary level dealing with farmers. Such instititions facilitate the provision of credit in kind (and on time) and the collection of repayments by deductions from the proceeds of marketed produce. They are also better situated to pursue supervised credit provisibns; to ensure its purposeful use; and assist in recovery. Further, they allow the availability of credit to be tied in more effectively with the government extension services to become a vital catalyst in the process of technological innovation. While in practice there is probably a need for both specialized and multi-purpose credit institu- tions, the advantages of delivering credit through the multi-purpose type justify considerable effort toward resolving the credit management problems that frequently crop up in such instititions.. 30. The best means of reaching large numbers of small farmers is still unclear. Most Bank Group l'ending for farm credit has gone through commercial banks, via the central bank. This has proved satisfactory from a management and control standpoint, and has been effective in reaching the clientele of large borrowers. In general, however, such institutions are limited in;dealing directly with small farmers, because of high-administrative costs, lack of borrower collateral and locational limitations which restrict access. Govern- ment efforts to prescribe quotas for lending to agriculture, or partial guarantee of loans to small farmers, have not moved the commercial banks substantially in this direction--despite serious attempts to do so by the banks in some cases. Agricultural Banks and Development Banks are bound by the same kind of limitations, though they len,d almost as much to agriculture as the commercial banks. 31. One avenue of lending to small farmers - sDecial crop and Droiect authorities - has provided a satisfactory channel for reaching relatively large numbers of smallholders. Since crop authorities can provide a guaranteed market (perhaps with quotas) and are usually in a position to ensure Drofitable return. they have been able to onerate successful credit operations. The nature of these institutions facilitates the delivery of credit in kind, its timely dintribution to coinicide with field oneratlons, sunnart and super- vision by technical field staff, and the collection of repayments by dedictionnsz frnm irpti-,nas G1ler1V th4Q _sQ an pffprtiup mannS nf dil iuprlng seasonal credit to smallholders, and should be utilized wherever possible. An.JT importannt qualificationmight hp tht atiish cre it sheuilA nno nnly relate to the needs of the cash crop being produced, but also to the subsistence crops grown for the farer's or. use, with repayments for both hb4nrg deducted from the cash crop proceeds. Credit administered by project authorities as a ,z.ponmnt of 4r.tegrated d Velolentor. schemnesn hasa been aa -o-efull man.s -------- -. - I ,- -..50 *meal._ of reaching substantial numbers of small farmers in limited areas - though t nheA 1 1 overal l eor has be oA Tn *-h L.TATTT ---A 4,. (P4A1,4 -4.__ACn4.._ .flt. J V . I. ak..' A CL.%J ALlX **flO.J..LA* A.-. [ *A V& J jL JL-, j-. tion of maize rose by two or three times over three years. On the other h.anA Jin I lJ-Iongwe (Mal1aw there was Adeclir4. -.. production though t-he caus may have been unrelated to the project elements. This type of approach, trhougih lA.m4 t4- in t,he scale 4i- 4ar ha,, 4s p:obaly. Whm app:opriate An aon African context where suitable institutions and basic infrastructure are n I1A__.. An 1...,.L4n, ov-.-terwise -lacki-a 32). Cooperatives are allso an important elemnt cor reachir.g small .J4... ..fJ.JjPLO.Lv~ Imp OO'JU JA h uSA L . OJO AL. J.I J . &-AJA6 OLU farmers. Almost 20 percent of Bank Group farm credit is disbursed through cooperatives-. --k.es- uayav an apex structure, layered tU th lo or primary society level. India has channelled a large part of its agr4cultural credit t-hrough the- T-ar. lev--I, BC.opeaie yt,-- which has performed fairly well. Successes have also been recorded in Taiw.a4 -.d-A Voa A, in 4isolated 4istances in -many other coun.ties. But .L.WULI OAAL &~LV.CLG, OlLALL A. L U.LaL L .A.LALLAtheAU~L7LFA~ L~MALJ.~@.J many experiences elsewhere have been discouraging in terms of high adminis- trative costs, delir.quer.cies ar.nd inadequate ar,d ur.;imelly Tevcs Local LL~.L.LV~ t. aLa, ~.L±LA u~LAL.L~O OALA .LAaLA~4U~L~ .LOWLULJ.II'.L7 L. "D'C.J.d cooperatives are often captured by the well-to-do, and smaller farmers have difficulty 1in obtalning funds. A highly centralized sytem Udoes UnL necessarily ensure better performance than a flexible village society - 12 - structure where the operating decisions are made at the local level. In general, the proficiency and equity with which the cooperatives operate reflect the conditions prevailing in the country as a whole - a fact which is hardly surprising. Where there is a strong private sector, cooperatives are difficult to sustain. They also encounter serious problems in countries which have an acute shortage of skilled manpower. Nevertheless, cooperatives probably provide one of the most promising vehicles for reaching large numbers of smallholders and rural producers. 33. There are other means of reaching small farmers. These include farmers' associations and other groupings, frequently informal, which have obtained credit from financial institutions under conditions of group lia- bility. Small groups have been organized, sometimes within a formal cooperative system, in several countries in order to obtain credit for small farmers. These groups are highly diverse, including village societies in Turkey, tribal groups in Africa, peasant groups in Mexico and so forth. 34. In areas where credit institutions are lacking or grossly inadequate, it is possible, as an interim measure, to increase the supply of credit for small farmers by using non-institutional channels, such as traders or merchants, as conduits for public credit. This has yet to be proved as an effective mechanism. The risk of misuse of funds and the difficulty of integrating credit functions with the other package elements indicate that the use of non-institutional channels should be approached very cautiously, usually in the context of a channel of last resort. 35. There is need for all concerned to learn much more about the most appropriate channel for providing credit at low cost to enable large numbers of small farmers to become more nroductive. It is clear. though, that any system intended to reach large numbers of low income producers will have to hp hased on different nrinrlnles from those syvtems assigned to resah relatively few large producers. One important difficulty which arises from the needi to redptr-e administrat1ive costs is thnt the horrro.wers themselves will have to operate through some institutions which represent them as a grour. This institultion w-ill have to asme respor.sibilty fnor heingr the channel to its members and for administering, supervising and collecting loans at the local level. However, while it is possible to identify the need and importance of a group for this purpose, the specific nature of the group that will be monst effec 4n t vive will vnr a n A d4ffrer.t nn*4 mand cultures. 36. Because of the special problems of dealing with small farmers, a case could bDe made for separate cred A4i nstItutior,s tn o lookl after them. They require much more service, including closer supervision, than the more coUU..ercialized larger farruer an.d more flexible poli5cies relatin.g to credA4t collateral, down-payment requirements and repayment schedules. But insti- tutional anrd manpower constraints are such that few countries are in a position to establish parallel institutions. A practical solution is to eep tiL'e prUoLgra adIU accUUnLts LU1 sma *dL LW. LUCLW aU1LC U U.LO-U%L- from the others, so as to be able to evaluate performance and costs of - 13 - providing credit to small farmers by various methods, and to take such corrective steps as may be required. For the same reasons in a multi-purpose cooperative, the credit account should be held separately from the other activities. 37. To sum up, in order to develop effective credit institutions for dealing with small farmers, these appear to be the important considerations: (i) The institution must encourage acceptance of its role in assisting small farmers and make itself readily accessible at the village level. (ii) It must view credit as part of a package to improve small farmer productivity, have specific proven technology to do so and assurance that the inputs required are available. (iii) The institution should look to the advantages of providing credit in kind for purchased inputs, both to relieve the smallholder of further transactions with which he may be unfamiliar and to provide the institution with some assurances that the credit is used for the purposes intended. (iv) Credit, especially credit in kind, must be timely; if provided too early or too late it leads to diversion and loss. (v) The basis for selecting smallholder borrowers should be creditworthiness but the criteria need not be as restrictive as for larger borrowers. The important elements should be the reputation of the individual within the community, the technical feasibility of the proposed enterprise in his own farm situation and the expected cash flow generated. (vi) The prospects for repayment of loans are greatly enhanced by group responsibility for individual liabilities. Given the cohesiveness of most rural communities, when the village cooperative society or farmers' association has a stake in an individual's performance, it is difficult for him to withstand the pressure of his peers and avoid his obligations. (vii) Institutions should understand that for small farmers, more than for larger farmers, loans and repayments need to be carefully scheduled to meet periods of liquidity shortage and surplus as they arise. This will make supervision much more effective and orderly. (viii) Even so, institutions will need to exercise considerable flexibility in rescheduling repayments when unexpected circumstances, such as drought or other disasters, occur. Under such conditions, it may also be necessary to be flexible in regard to lending criteria. - 14 - (ix) There must be a commitment on the part of the institution for contiLnuiLty oU operatiLons. IsLL iL8 d LaCUcgLL.LLOL1 that performance in the initial stages may well leave much to De desLred. it will take time and discipline to develop effective credit programs for small farmers. (x) Finally, authorities should recognize that there is still much to be learned about small farmer credit, that a process of trial and error on a limited basis may be quite in order in many circumstances to provide guidelines for wider application. Ultimately, the program should be conceived as providing continuing and increasing financial support to the farmer for the evolving process of modernization. Recommendations 38. (i) The Bank should continue to expand its efforts to help meet the credit needs of small farmers. Increased attention to small farmers should not, however, obscure the need for aL;o in- creasing significantly the aid to other groups especially medium-sized farmers. Reaching large numbers of small farmers in low-income countries with credit, at the farm level, will necessitate a substantial volume of Bank/IDA lending. During the next five years an anticipated US$1.3 billion will be allocated for rural credit and close to half will be for small farmers. (ii) The expansion of rural credit should be on a selective basis related to situations which will lead to an increase in rural productivity. Credit is not normally the appropriate instrument for transferring income to low-income producers or the rural poor. While an expansion of credit could reach large numbers of small, potentially productive farmers, there will be large numbers of low-income persons in the rural areas, such as the landless and those whose holdings are too small to be viable who will seldom benefit from rural credit programs. They will have to be aided by other means. (iii) Access to short-term credit for purchasing fertilizers, improved seeds, pesticides, etc. is often of greater importance for small farmers than long-term credit in the initial stages of the transition to more productive agriculture. Accordingly, there should be appropriate emphases in the credit programs for small farmers toward short-term seasonal credit in the context of overall on-farm development planning. Bank Group lending could provide a permanent working capital fund for this purpose which is rolled over and re-used every year. As the Bank Group loan/credit is repaid; domestic sources of credit/capital can gradually replace foreign ones. - 15 - (iv) Emphasis should be on the productive capacity of small farmierc rathpr than nn rnllatsrn1 as a crit-rinn fnr 'Inni decisions. For long-term credit to large farmers, land mortgages are in order; mnre use shoiuldA ha made of chatte-al mortgages or liens on crops as security for short- and medium- te.m. loans . In geeral, further e -ermon-sto 4 4#-1i simpllfied arrangements and procedures to secure credit should be encouraged. (v) Preference should usually be given to a single institution h-anAl-ing -1 types of .redit neeAeA b-y farmers, b-otlh short-term credit for production and for living in the 'Lear periLodU as we.LJl as meudlu.um- auld long-term credit Lor on-farm development. This would benefit the recipient an cu sLLdU~ LLLC .LLf L.. L ULLUIL LU VVCLD *. LLD IU protect its loan. Credit to farmers should be viewed as an rLLtegral part of a continu-ing process to heLpL' f amer adopt changing technologies as they evolve. (vi) With regard to small farmers, preference should be given to projects il WIbiilCI creuiL iS UdeigIneU au a element oX a reasonably low cost package that provides the necessary additional services - as in integrated area or crop develop- ment projects. (vii) The Bank should encourage appropriate adjustments to the general structure and level of interest rates so that inter- est rates within the agricultural sector are in harmony with those in other sectors of the conomy. In general there is no simple or unique answer as to the question of what is an appropriate interest rate on a Bank financed agricultural credit project. The Bank should work toward tie long run objective oL all interest rate that reflects the costs of capital and of providing the capital; an intermediate objective might be to cover at least the opportunity cost of capital. Subsidies should, in general, be limited to cases where L1' cani be Justified d are Like'ly to be effecL ve in view of the pattern of faraing in the project areas. The issue of interest raltes, however, should always be seen as one corupouLent in a project. The extent to which the project would achieve the overall objectives of Bank policy should influerce the attituAe A-o-sar interest rates. (viii) The Bank should encourage governments to develop systems of lending through cooperatives and/or groups of small farmers. This approach seems to hold most promise for reaching large numbers at low cost. The nature of the groups will vary according to the culture of the different societies involved. At the same time, it should be recognized that cooperatives perform poorly in most countries and ways need to be found to strengthen them. - 16 - (ix) Special project authorities such as those created in Africa in limited areas should continue as Bank instruments to draw farmers from subsistence agriculture. At the same time, the Bank should encourage the development of apex institutions which can promote, oversee and service such enclaves within national programs. (x) The Bank should give increasing attention to building and strengthening financial institutions as agricultural credit channels, particularly to orient their efforts towards the small, farmer. 39. Clearly, the Bank needs to know a lot more about how small farmer credit programs (including non-governmental programs) perform in improving productivity and incomes, and the costs involved in providing such credit. Accordingly, the Bank Group should in fucure insist on a reporting system being incorporated, at least in major projects, which will provide information from the farm level-regarding the situation before the project was initiated and the progress and problems that emerged thereafter. In the. meantime, the Agriculture and Rural Development Department should initiate a study of the administrative and other costs. of providing credit to small farmers. This study should include considerations relating to the effectiveness of cooperatives and other group activities as a means of providing low-cost credit. I. AGRICULTURAL CREDIT PRACTICE AND PROBLEMS 1. Evolution of Current Approach 1.1 suThe goals of public agricultural credit programs have chaged sig- nificantly over the last two decades. Historically, the main objective was to reduce the farmers', especially the small farmers', dependence on tne vil- lage rwney lender, who, it was thought, exploited farmers through usurious interest rates. In addition, refinancing was needed for many farmers, espe- cially in Asia, who were excessively indebted at high cost. However, with the emerging importance of economic growth in tne 1950's and tne development of neii and more productive agricultural technologies in the 1960's, govern- ments have shifted emphasis so that today the main purpose of most credit programs is to increase output. In a few of the more prosperous developing countries there has been a further evolution: credit institutions are now used to mobilize rural savings in order to reduce the sector's dependence on external funds. The programs receiving support from the World Bank Group and other ini:ernational agencies have increasing production as their primary goal. Many combine this with a concern for the welfare of the small farmer, again primarily through raising his output as well as reducing his dependence on the money lender. 1.2 In idealized form, a modern, production-oriented credit program is organized as follows: the government lends its own funds together with those obtained from other sources (e.g., an international agency) to an agricultural bank which, in turn, relends the funds either directly to farmers or indirectly through coooperatives. The farmers use the funds to purchase productive in- puts - fertilizer, seeds, pesticides, livestock, tubewells, machinery, and such - which are combined with family labor to produce more output. The addi- tional output is sold and the proceeds are sufficient to repay the loan yet leave the farmer better off. The payments received from the farmers by the agricultural bank are adequate to cover administrative costs, to pay the interest on the government loan and also to regenerate lending capacity. 1.3 The above paragraph suggests several criteria by which to judge credit programs: their success in increasing production and farmers' incomes, their success in generating sufficient interest and repayments to meet insti- tutional costs, and, for those programs with a small farmer orientation, their success in actually channeling credit to large numbers in that'group. It is difficult to provide "hard" evidence on the impact on production but it appears that few credit programs can pass all three tests. In a number of countries the availability of institutional credit has undoubtedly helped to increase output, but the impact has not been quantified. Regarding financial viability, with few exceptions the record is poor. Interest receipts seldom cover costs and many programs have high overdues. Without substantial subsidization, few programs could survive. As for the distribution of credit, most of the insti- tutLonaJL loarns Lh ave gone to larger farumers. Furt hemUL ore, a 'Lew credLit pro- grams may have been damaging to the position of the smaller farmers; for exam- ple, in Colombia, Ethiopia and "akistan, the new technology financed by loans contributed to the displacement of tenants (due to the lack of concomitant changes iLn inst'iutiLoUnal U LLUCLJUL, tesJj;;ei.Lai.Lly LanLU LteUre). 14 Credit is aii impLportalt tooL inLL 'osterLing development. But, uniess the recipient groups have profitable opportunities in which to invest, unless the programs are well designea ana aumiLnistered, and unless great effort is made to hold down the rate of delinquency, credit programs will fail to meet one or more of the criteria of success listed above. Tnis paper attempts to outline the issues associated with agricultural credit, particularly with lending programs for small farmers, and suggests policies for alleviating the problems. Volume and Nature of Institutional Credit 1.5 Outstanding institutional loans for agricultural credit in LDCs are estimated to be in excess of $15 billion. international support for these programs is of considerable importance. The World Bank Group is now commit- ting each year approximately $400 million for agricultural credit of which more than $300 million is on-lent to farmers; the Agency for International Development and the inter-American Development Bank each supply another $50 million per year and other assistance agencies provide further millions. 1.6 Institutional lending to the agricultural sector by country is de- tailed in Annex Table 1, which shows total loans outstanding and new loans made by institutional sources in the most recent year for which data are available. The table also gives the level 6f institutional lending in rela- tion to the rural population. There is substantial variation among countries: in a number of countries, loans from institutions are less than $5 per person living in rural areas while in five the amount is in excess of $100 per cap- ita. On a continental basis, in most countries of Africa and Asia institu- tional lending is less than $20 per rural inhabitant, while in many Latin .American countries the amounts are in excess of $50 per capita. The greater importance of institutional lending in Latin America is confirmed in Annex Table 2, which reflects the distribution of loans by type of lender. Again the figures vary widely by country, but, on a continental basis, the percen- tage of loans being made by institutions is high in Latin America, while in Africa and Asia non-institutional lending predominates. Within the institu- tional sector, public institutions are of much greater importance in Asia and Africa, while in Latin America a substantial fraction of loans pass through the commercial banks. 1.7 The percentage of farmers receiving institutional loans is shown in Annex Table 3. Again there is a great difference between certain African countries where little more than one percent of all farmers use credit, and Taiwan in which nearly all farmers have access to institutional credit. On a continental basis, about five percent of farmers in Africa get institutional credit. while coverage in Latin America and Asia (excluding Taiwan) is about 15 percent of all farmers. 1.8 In summary, only a relatively small fraction of farmers receive institutional credit today. The remainder either do not borrow or are depend- ent for loans on money lenders or friends anid relatives. Though institutional credit is growing rapidly, non-institutional sources are still the major sup- pliers of credit to farmers in most less developed countries outside Latin America. 1.9 Coverage by institutions is even more limited in the case of small farmers, for in almost all countries institutions have channeled their funds to the larger farmers. In Pakistan the smaller 60 percent of the farmers got three percent of the institutional credit. In Bangladesh, few farmers hold more than three acres and yet these larger farmers captured more than 80 per- cent of the loans from the Agricultural Bank and the cooperative banking sys- tem. In the Philippines 27 percent of the larger farmers handling 61 percent of the land captured 98 percent of the institutional credit. In Thailand, those receiving institutional credit held, on average, 60 percent more land than the average farmer; in Tunisia, 90 percent of the farmers could not qual- ify for institutional credit; and in Bolivia only 3.5 percent of the institu- tional credit goes to the campesinos. In Brazil three percent of the farmers got 34 percent of the loans. Studies of Chile, Colombia, Ethiopia, and Hondu- ras - all indicated that at the time of the survey the larger farmers were the main beneficiaries of institutional credit; this indicates that there is scope for expanding institutional credit to small farmers. Development of Bank Group Participation 1.10 The Bank Group's lending in the field of agricultural credit has eo-panded rapidly, and now constitutes the major part of Bank Group lending to agriculture. 1/ Total Bank Group commitments to agriculture rose from $468 million in the sixteen-year period Fy 48-63, to $621 million during the five- year period FY 64-68, and showed a more than four-fold increase to $2,589 million in the most recent five-year period (see Annex Table 4). The credit component grew even more rapidly, rising from less than 20 percent of total lending to agriculture in the first period to 56 percent of the total in the last five-year period. 1.11 There has also been a significant increase in the share of credit funds going to countries with lower per capita incomes (as also seen from Annex Table 4). In the earliest period of Bank lending for agriculture to countries with per capita GNP of less than $150, less than four percent were for credit proiects. In the most recent period, agricultural lending to the poorest countries exceeded $1 billion of which more than half was for credit. Clearly, credit has become an important element in the Bank Group's agricul- tural programs for low income countries. 1/ The data discussed below refers to all projects in which a minimum of 10 percent of the Bank Group loan was used for agricultural credit purposes. While this creates some inconsistency with earlier Bank documents, e.g., the Agriculture Sector Working Paper, the broader coverage provides a more accurate measure of the Bank Group's involvement in agricultural credit activities. - 20 - 1.12 Tnis activity in the low income countries is of course directly due to the availability of IDA funds (see Annex Table 5). While negligible in the earliest period, by FY 1969-73 IDA credits were utilized in nearly three- fifths of all credit projects and provided over two-fifths of the volume of lending, not counting the portion in projects which were a Bank/IDA blend. 1.13 The Bank Group's initial credit activities were guided largely by three concerns - that funds loaned should lead to increased agricultural production and productivity, that the investments financed should constitute an economic use of resources for both the farmers and the nation. and that support is given to developing agricultural credit institutions. In practice this meant a concentration on commercially viable farms and agriculture-related enterprises; credit was not regarded as a practical means for dlealing with the nrohImq snf .subsistence farmers and agriculttural laborers.- However; the technological improvements of recent years have changed the economics of sma11-sqr1P far-mingj making it nnssible for onne marvinal farmn to beromne viable and creditworthy enterprises. The Bank Group has been increasingly concerned with this group of nrodu_irrers, and 21 though the Bank contiJnwes t-n assist programs lending to medium- and large-scale farmers, there has been significant shift in emphasis in the allocatIont of its resniures to these small farmers (see Annex Table 6). Approximately one-quarter of credit financed by Rank!/TA is inteneipd for smnall-scale producers. 1.14 In proJect appraisals, it was reported that agricultural credit operations during FY 1969-73 would benefit over 900,000 small farmers with holdi ngs of fi4- hect ar or Tles In nAAd4dt4io. 300 cooperatie wre bene- ficiaries, containing small farmers as well. Almost 80 percent of the farm- ArC hol 44n less than f4im hen*trne AOnA b0i0F4J t4- fra, -T^A" , r-. F.-nAc ..- in India and Ethiopia. Bank Group credits to India go to 300,000 smaLl farm- ers , 'about 0. 7 percent of tChe estl-imated 42 1)milli -on smallholders in 4 th' coun- try. The loans are mostly for wells, pumps and motors and have an average size of approxi,a-telty $1000, _of whichthe B.k provides about $650. Tlhe - .kt commitment to these loans is $185 million and constitutes 65 percent of the volum.e of a..l Ba.. k moe.ey flowing tCo small fa..,..ers in thlf.-e fo=. of loans. Another 400,000 small farmers in Ethiopia receive Bank assistance as part of thLe Agri.u`luraU Y1.nlmum laciage ProJe.t. The average loa. size is approxi- mately $25.00, of which $12.50 is provided through Bank lending. These loans are for-. seasoUna AnLpULt such as see's and fertilizers. I'se Lnu-bI Ler olu sWa0Ll farmers benefiting from Bank-supported programs has grown in recent years, but at best are reaching only onLe perCent U' LLIth IL0 IUL.L.iorL mUaLl aLLUr in the developing world. 1.15 The country and regional pattern of Bank Group credit activities (shown in Annex TaDle 7) brings out the prominent role Latin America has played in past Bank Group credit operations. Only in the latest period, FI 1969-73, did Asia approach it in volume or lending and number of projects. The table also reveals the Bank's limited involvement in agricultural credit in West Africa - a reflection of the difficulty of mounting credit projects in that Region, in which traditional cultural practices predominate, and the fact that many countries did not achieve independence until rairly recently. Two countries, India and Mexico, together received nearly two-fifths of all Bank Group credit for agriculture during FY i969-73. - 21 - 1.16 A review of the contribution made by Bank Group financing to the total costs of agricultural credit projects (shown in Annex Table 8) reveals the significantly higher proportions of project costs provided to low-income countries - a position which emerged most clearly during FY i969-73. Also, local cost financing has tended to be more important for low-income countries than for those in the highest bracket. By the FY 1969-73 period, countries with $150 GNP per capita or less had, on average, nearly two-fifths of local costs covered by Bank Group financing. Ability to do local cost financing has been a key element in expanding the scope for Bank Group participation in agricultural credit, particularly in those countries with the lowest in- comes and at the earliest stages of development. Categories of Bank Group Lending 1.17 Information on the types of credit projects financed by the Bank, as presented in Annex Table 10, shows that total lending in each category has increased in the three periods covered. However, the rates of expansion by project type and by region differ considerably and are indicative of changes in the Bank policy. Lending for livestock operations remains the single most important type of credit operation, though its relative share in Bank credit commitments has declined slightly. Livestock loans constitute nearly one-third of Bank credit projects and make up more than 70 percent of the credit projects in Latin America, which, in turn, has been the region receiving the largest number of Bank Group credit loans. Within Latin America, in the past at least, most of the funds have gone for the development of large- scale commercial ranches. The trend in recent years in Africa and the Middle East, and in Latin America as well, has been toward developing smaller scale livestock operations. 1.18 The second largest category of loans has been credits in support of general agriculture. The nature of such prolects is quite varied, the category including loans for mixed farming, for general on-farm improvements such as land-levelling and for the purchase of non-mechanized implements, such as plows. 1.19 In the period 1969-73, agro-industrial credits were third in impor- tance. These are classified by the Bank under agricultural loans but are not for crop or livestock raising; rather they are for processing plants, storage and marketing facilities, and for aerial soraying of crons. The next class. in value of Bank credit operations, are loans for minor irrigation projects. 'lost of these nroiects are for wells and low lift numps. In recent years; this type of operation has been limimited to Asia and is often to provide the water control essential to the adontion of the new varieties of wheat and rice. 1.20 Lending for crop development, mostly export crops such as tea, oil palm, and cocoa, and integrated agricultural development have both emerged as important credit activities in the 1969-73 period. Reliance on these types of projects in East and West Africa reflects the specialization of commercial agriculture in these regions, and the need to package credit together with extension and infrasturcture in Africa. - 22 - 1.21 Farm mnrhinery loans were one of the Parlvy kinds of agricultura credit activity in the Bank. These frequently involved financing imports of mTrhrner, and t-he us, e ovf the counternprt fitndt frnm their co1c forv orle.diA4r. to farmers. Though lending for this purpose has continued to expand, it has done so at a slower rate thaT. lendin- for other purposes and is now concentrated in Asia rather than in Latin America. There is awareness in the Bank of possi- ble adverse social effects end the dislocation of labor associated with too rapid mechanization of agriculture. 1.22 Agricultural credit institutions tend to be treated in isolation ratiller tLan as a segment ofL the rural fin.ancial system. mTh uiscussion pre- sented above indicates, however, that institutions provide only a fraction, and in rmost countries onrly a small fraction, of credit used by farmers. If farm savings are included, institutions provide an even lesser part. A recent stuUy i1n lllUnLda reported' th'Lat over hI allf thle farm households diad not nave any debt at all; farmers that did borrow still financed 95 percent of household expenditures, 9§0 percent oL operating _uly an 50 pecent of. investmen- costs out of their own resources. Of the funds that were borrowed, only 30 percent came from institutions and instLtutLons rinanced only one percent of total farm outlays. Institutional credit programs must be designed to operate within the overaii rurai rinance situation. 1.23 Some of the major problems in rural finance stem from the fragmenta- tion of the financial markets. In some countries, funds do not flow readily, either between markets or even among borrowers within a market. Where this situation exists, it can lead to the following problems: (i) a lack of com- petition among lenders, leading to usurious interest rates; (ii) an inelastic supply of funds, slowing the investment, process when the demand for funds for development is increasing rapidly; and (Iii) the use of surplus funds whether for consumption, land purchase (exacerbating the land tenure problem), or for other investments not related to increased productivity. Each of these prob- lems is basically associated with a different stage of agricultural development, though a monopoly of credit supplies, where it exists, can be a problem at any stage. Moreover, because the farmers of an area do not all pass through the stages simultaneously, all three may be problems at the same time. Limitations of Informal Sources 1.24 In traditional agriculture with a stagnant technology, the output of farmers is stable or expanding slowly, and investment is low, not because farmers are too poor to save or interest rates too high to borrow, but be- cause, over time, they have acquired the quantity of capital 1/ which is con- sistent with their technology and their holdings of land and labor. In a 1/ Tn this paper thlIe termLcapital' 'Ls usedU to refer to fixed and working capital, while "credit" is used to mean a transfer in-cash or kind with an obligation repay. - 23 - traditional setting some poor farmers use credit regularly to pay for ;their household requirements in the months bEfore harvest. But most farmers borrow only when their crop has been poor or they are faced with an unusual expendi- ture. often for a family ceremony such as a birth. wedding or death. 1-25 Muich of the borrowine is from other farmers - neighbors. friends, and relatives - who charge a nominal rate of interest but expect comparable finanning shnuld they find themselves in need of credit. Some farmers. and especially those small farmers making regular borrowings, obtain loans from mprchantqs T,iRA1pmePn iand money lenders; who eharae high interest rates. Tn some places there is competition among such lenders and the rates charged are rougxly eqiiuilent to the high rnst of lending to small, rural borrnwers, including a realistic risk premium and the opportunity cost of the money lender funds6 Other money landfvr have A monnnnno v lnnstion and Ara Ahlp rn rWh:ro rates greatly in excess of competitive market levels. 1.26 In those countries of Asia and Africa for which figures are avail- able, friends ad relaties p 50 percert of total I_r.s; for t countries in Latin America, they provided only 10 percent (see Annex Table 2, ttm- - 4.1 ltav.A,a A. Q naQh 1^1nna vntvact -a n% rt gf- pn_aAvt 4a n-.4. nal. Still, some farmers, and a disproportionate fraction of these are prob- ably sal,zll farmerts, are forced to borrow from mon.e- le….ders, …iddlem-, lnd- lords and merchants, who, on average, provide 27 percent of total agricultural creAuitL (see Arl.ex Ta 2Ul ., 9 UUUmercaUt lenders). Man o qthe 10- fro th commercial sources are at very high rates of interest. Particularly in Africa tAhe rates on commercital- loa..s are v-ery high (see Annex Tatble ON, .u. t- the v- JJ of commercial lending there is ve!ry small. The interest figures may be mis- leadir.g since- they rLepr~eser.t ar. ar,nuallizati4or fmrtl ae,adms on at high rates are for short duration, seldom longer than three months. 1.27 Exploitation is made possible by ignorance, by poor communication, by thie ab)senice of alternative lenUders, buy established tra'rng patter.-s ar.d by differences in economic and political power. Public credit programs by orfering an alternat'-ve source of IunUd caLl Lltp LtUho Who -mUsL Udal WLIL a credit monopolist. But monopoly power may not be the only, or even the most ±mportatLL,L rteasoULn for iLgh.1LI -iLt.erUest rat=es On agricultu.0 LS8 in deUVCl0.iUg countries. Ratner, rates on commercial loans could be high in part because of monopoly, but also because capital is scarce, because farmE lorLIs are c;1ULy to administer, because the rate of default is high, because much of the demand for credit is seasonal, and because in many couLtries there i8 substanltil inflation. 1.28 The following illustration suggests the level of interest rates wrhichi could be achieved only by the most efficient private lenuers, namely - 24 - lenders wiho could invest their capital during the whole year, 1/ and who could hold administrative costs to three percent and defaults to only three percent. To earn 15 percent in real terms, they would have to charge a real interest on loans in excess of 21 percent per year. 2/ This is a competitive non-monopo- listic rate. On a monthly basis it is about 1.8 percent per month. Yet even at this high rate, the cost to a farmer of financing half of his total expendi- tures for four nmonths through borrowing would be about 1 percent of his annual expenditures. Data on India indicate that few farmers actually borrow this much. The interest rate figures on loans by commercial lenders show that in all Afriea and in selected countries of Asia and Latin America. the rates charged by money lenders are in excess of 32 percent in real terms. In these countries; there may be some mononolv in the credit markets: however- shorter loans, higher administrative costs or more delinquencies than assumed in the anhove Px,mnlp wnuildi i,qtifv hlgher -omnetitive ratas. Tn 9re other ciountrlesa the money lenders' monopoly power 's apparently quite limited. 1.29 Only a few studies throw light on the question of informal market inteprest rates.- Theos indicat-e that in Eadr,rr Tndian TnAnnesia Thailand and Vietnam average interest rates on commercial loans are not out of line with competitive rates. Stuf ... 4a1n Maalir and Ch4le however, inrdicate that in those two countries there is a substantial component due to monopoly inlcl_de in the interpet -rate charged by money lenders. 1.30 The object nof fru ^mnt olicy sol be% A ho toel1r4n nta nont the money lender but his monopoly profit. The most effective way to reduce interest ra.tesl nad aa.nl,.4tatln 4a t-- 4",-nnaaa hlrt-h ~t-hc-nltav- t4Jira cn,.¶--oa anA 4-ha= zates ~ an exliaion. is to ir.crease both- the alternative sores.- h volume of credit in agriculture. This can be-done either by establishing .oe.flrn¶C,f credit agencies or by pursulng policies that _ourage priv_at institutions to lend to agriculture, by allowing them to charge interest rates high enough to r,ake lenuding to famers profitable. I/ I1-Mrchiant-lenders usually invest tL LleiLr capital iLn loans over a periLodU of the year. In seasonal agriculture, there is in effect a symbiotic switch- ing over the year oL worki..g capita'l between farmer andU merchLants whL to are crop-buying agents. The merchant requires substantial working capital after the harvest to carry out the crop. over the year as ne sells out his position, his need for working capital is diminished, freeing funds for 'Lending just at tne season t[hat farmers are running low on money and need loans to purchase household supplies. This seasonal shifting of funds between merchant and farmer makes for an erficient use of capital over the year. 2/ The calculation is as follows: to have an annual return of 15 percent on his capital, the lender must collect 15 on every 100 lent plus 3 for administrative expenses and 3 for losses - that is, 21. This must be earned from the actual capital recovered - that is, 97 over a period of 12 months. The lender must therefore charge a real rate of about 21.6 percent. - 25 - 1., 31 I nLUily govermerts atteUpt to LULIU.L6 L&&= LtLh Uat of L,tesLt chLargeU by the private sector through usury laws. If such laws could prevent exploita- ti.Lo, thLey wuuld be -must desirable. BuL m Ly LLenders a-LL.L Uorrowers do r.o; pay much attention to usury laws. If the maximum interest rate is set too low, usury laws discourage lenrdig to agriculture by those private istitutions, such as commercial banks, who usually obey the law. It is important for develop- ment to improve f'nancial markets and LU ±incres LtLh 'loW Uo LcUrdL, both between and within economic sectors. Financial policy must pay due regard to intersectoral interest rates. Limits on interest charges which impede the flow of funds into agriculture by institutions, but are unenforceable as regards the money lender, do little good. unanging Credit Uses and Requirements 1.32 Modernizing agriculture necessita es the introduction of new techno- logy into an area. The change can be of various sorts: it may be the intro- duction ot a crop not formerly grown in that area or it may be the adoption of some practice which preserves yield levels, such as soil or moisture con- servation, pest control, or perhaps the introduction of crop strains more resistant to drought, heat or cold; or the innovation may involve a yield- increasing technology in the form of new varieties of seeds, pesticides, fer- tilizer, or machinery. 1.33 The various types of innovation have different financial requirements. Some innovations require little additional capital; for example, a switch from one variety of seed to another can be made without much investment. The addi- tional capital needed for others, such as the introduction of pesticides, can be financed by many farmers from savings. Evidence from such varied countries as Brazil, Korea, Pakistan, Taiwan and Zambia indicate that farmers have sub- stantial savings capacity when they have attractive opportunities in which to invest. 1.34 But in addition to their own savings, borrowed funds are needed by many farmers, especially when the investment is large relative to their in- come stream. In modernizing agriculture, credit is less likely to be availa- ble from other farmers, as in all likelihood they will be using whatever surplus they have to finance their own investments. 1.35 Nor are local landlords, merchants and money lenders likely to be an important source of additional agricultural finance since their supply of funds is quite inelastic. They finance only short-term production credits; lumpy, longer-term investments almost always need to be financed elsewhere. Further- more, in the event of the introduction of land reform, this source may well disappear. 1.36 New technological opportunities can lead to financial stringen- cies when they require more resources than traditional sources are able to provide. It is at this point that access to credit from external institutions becomes critical. Without it, farmers' investment strategy and especially that of small farmers, will be biased toward marginal variation within the known, traditional technology. - 26 - Mobilization of Savings 1.37 One way to increase the flow of funds within the agricultural sector is to tap the surplus funds of those who have successfully adopted the new technology. To realize this potential, an appropriate savings transfer mecha- nism must be established. If credit (and savings) institutions can offer suc- cessful farmers attractive financial returns, they can encourage them to main- tain a high rate of savings. The resources so mobilized can then be re-lent to farmers who are still at the earlier stages of development. To the extent that agricultural credit institutions can become the foci of local financial markets by mobilizing and distributing savings, they can lessen the dependence of the agricultural sector on outside sources of finance. 1.38 It must be recognized, however, that mobilization of savings through financial mechanisms will not automatically lead to an increased flow of funds within the agricultural sector unless loan investment returns are commensurate with those elsewhere in the economy. Traditionally savings mobilized in rural areas have been re-lent primarily in urban areas (because of higher interest rates in these areas). 3. Constraints on the Role of Credit in Development 1.39 Credit - that is money - can itself grow nothing. To achieve the objective of expanded production, borrowed funds must be spent by farmers on physical inputs - fertilizer, seeds, pesticides, labor. The surplus out- put must then be transported to market and sold to domestic or foreign con- sumers. This is a complex process. Credit puts funds in the farmers' hands that can be used to purchase productive inputs, but whether this will be done or not depends upon technology, markets, infrastructure, information and attitudes. 1.40 In most developing countries, the growth rates in agricultural output have been the slowest of all major sectors: farm production gener- ally has been increasing at less than three percent per year. Coupled with this, there has been a low level of capital formation in the countryside. Many credit programs are predicated on the assumption that, in large part. a shortage of funds is responsible for the slow rates of investment and growth in agriculture, particularly in the case of small farmers. The follow- ing section aims to identify those factors which are essential complements of credit in promoting agricultural development. Limited Applicability of New Technology 1.41 There are many opportunities to put credit to productive use in agriculture, ranging from merely spreading fertilizer where none was used before; to utilizing the more advanced elements of the technologv assoriated with the full-scale "green revolution" including multiple cropping. But these are often limited to farmers in saecific regions or to areas with particulnr natural endowments. Although dramatic increases in yields were experienced after the successful introduction of new technology in some areas of the world in the 1960's, the applicability of these practices has been geograpnicaily limited. Indeed through 1970/71, outside of Mexico, the "green revolution" was highly concentrated in Asia - South and West Asia for wheat and South and East Asia for rice - with small quantities of the higher-yielding varieties of wheat being raised in North Africa and rice in Latin America. In fact, outside of Mexico, 86 percent of the total area planted to new wheats was in India and Pakistan. Rice was not quite so concentrated but 60 percent of the land planted to the new varieties was in India. 1.42 The possibilities of introducing high-yield and/or multiple-Jcropping techniques are limited, at present, to irrigated and high rainfall areas of mild temperatures, which probably do not occupy more than 30 percent of the world's potential arable land. For many other areas it is possible to develop yield-preserving, rather than yield-increasing, technologies, but there is an urgent need for more basic research to open up new technical horizons for agriculture in the developing countries. Similarly, there are limits on the availability of uncultivated land. Although there are areas of cultivable but unused land in Latin America and Africa, and a little in Asia, the expan- sion of cultivated land in developing countries is, on past performance, unlikely to exceed one percent per annum, a rate far short of population growth in the developing countries. Economic and Social Constraints 1.43 Even where technical opportunities exist, these may not be economic. New grain varieties, although they may produce larger harvests in physical terms, have sometimes sold at a discount because local consumers consider them inferior to native grains. The profitability of new investment oppor- tunities resulting from new technologies or from opening new lands is clearly not guaranteed, and it would be misleading to assume that all new technologies made available 'o the small farmer will be profitable to him. Where new tech- nology does not exist, priority should be given by government to developing opportunities which will make possible increased output on economic terms, or, where feasible, to taking steps to bring additional land under cultivation. 1.44 Some of the recent improvements in technology involve indivisibili- ties which make them less suited for adoption by small farming units. For example, the new seed varieties are much more productive when water applica- tion can be controlled. Yet the minimum size tubewell or low lift pump avail- able in most areas is larger than that required by small farmers to irrigate their land. In some areas it has been possible to group small farmers to share a single pump or tubewell, but where farmers cultivate several tiny parcels, such organization becomes more difficult. On the other hand, fertilizers, pesticides and new seed varieties are almost perfectly divisible. In the Indian and Pakistan Punjab, where irrigation was already available, the new grain technology is apparently as well suited to small farmers as to large, and the new practices were adopted to roughly the same degree by both groups. IWhen indivisibilities are important, the new technology tends to be less pro- fitable for smaller farmers, giving the larger ones advantages in the market place. - 28 - 1.45 Many agricultural innovations are quite risky. For example, the new seeds show greater yield variation than the varieties they displace. Under ideal conditions, output may be twice as great or more, but under adverse weather conditions the new seeds may vield even less than the traditioaal var- ieties. 1/ Many of the traditional varieties have evolved overtime, or were consciouslv develoned through earlv research work, to Produce under wide ex- tremes of weather conditions. The risks associated with the new technologies may threaten survival; esneciallv among small farmers in marginal ecological areas who live close to subsistence levels, and this reduces the attractive- ness of thp new nrartines- 1_46 New agF{icultur21 pFarticeS may be disturblng to farmerst ctilture- traditions, attitudes and values. Profitable changes in practices may not- be adopted if they involve work ennsidered to be demeaning, or if agricul- ture is only a secondary occupation, with primary orientation toward non- a.gricultura-1 eMplo, vent, n-r if snoiptipe prnviAd sanctionn noainst nprnogrespsiv farmers. 1.47 The absence of an adequate marketing infrastructure may make invest- ment .profitable. A n.mber of cou.ntry studies - Ecuador, (hana TInA4i, anrd Malaysia - report that a lack of adequate infrastructure makes marketing addi- tional output unduly const1vy Pricep nnlpirie- jsuh as nnrernliatir ponnrt ex- change rates, export duties, or artifically low prices designed to favor urban consumers can also reduce the profitability of -arketing addition4 1 output. Where c:he delivery and marketing system and pricing policies are satisfactory, experience show4s that not onhly are credit operatirn.s gr-etly facilitated but credit requirements are not as large as they would otherwise be. 1.48 Adoption of the new practices may also be constrained by a lack oz inus FL c^-.e r.ew seeds- -r. pesticides may be available, but ferti= -. EUL V_LULA~ ~.&. ka" w k..~ L lizer may be in short supply. Because success of the new technologies depends on a uaLancedU appLiLcat.LL of several Jputs, t,he abser.ce of ar.y one U-Lay ad- versely affect the benefits to be gained from using the others. An investment iin a tubuewe'lL ruay LcaAl if thle faLLUL.er C.annot al4soV obltain rMuiredl ferti' Lizer. The small farmer is at a definite disadvantage in acquiring essential :1nputs if their suppiy i's ±limJLeU to a dL'egree which requLreU rationing. SteepLy ris- ing prices may keep limited supplies away from the small farmers, but, even wnere prices are controlled 'aIu sUmeties subuusidized), L[Lh wUALLu±Cr ana more influential farmers seem better able to capture what is available and to get the government to respond to their needs. 1.49 Farmers may be ignorant of the econiomic opportunities open to them, or they may misjudge the potential returns from new practices. In many cases, recommended procedures appear to be only imperfectly adopted. Farmers may use new seed but not fertilizer or use fertilizer in seed beds but not in 1; The new wneat varieties introduced into morocco from Mexico were found to be highly susceptible to rust - a problem which had never occurred in the somewhat different Mexican climate. - 29 - their fields. For various reasons, including the great effort required to reach the large number of small farmers, extension agents spend less time vis- iting smaller units, with the result that the latter are less informed about new practices. A study of small farmers in Zambia concluded that an educative process was needed with regard to both the potential commercial nature of farm- ing and the profitability of innovation. Non-Credit Policy Requirements 1.50 To be successful in expanding production, government policy must relax each of the binding constraints, whether it be financial or non-finan- cial. For example. where the constraint is absence of knowledge. the Covern- ment's program should incorporate some form of extension service; where it is lack of experience with a nartieular intut or cronp the program might include a subsidy on the input or a support price for the output to make the use of the innut or the arowina of the cron more orofitable. if the constraint is risk aversion, crop insurance may possibly be used to reduce the yield com- ponent of risk, or a supnort price may reduce the risk of nrice declines. 1 91 There are alon varionu rtehninues for shifting Tore resniures into agriculture. In addition to institutionalized credit programs, alternatives inrInirlp riittlno t2SYPQ irnn n-lr.itilt-irrD nnvrtf,1n:r1v Pvnn,rt lp-ripa for imnnrn - -_____ … --…,… O- _r - - - ing the terms of trade between agriculture and other economic sectors. Many countries AisQriminate hbet-ween the prics eatablishea fnor agricutural ndA industrial commodities by raising the prices of the latter through protective tariffs and depressing those o the rmer by price controls on domestic food- stuffs and through taxes on export commodities. In countries pursuing such n0l t4,00q 4_n__-n,,in 4-0 i4-a-.., -C 4--nA_ C- So.-4nv14-ne h..lA U_ -4-,an 1h4_1-# r _knses, .proving the ten.ms of t_ d - or a-gri, ------ be given hig.. hest priority and would probably have a very substantial impact on output and con- sequently on the demand for cretA4-. fY.owever, it mmna.st bDe recogi.izA that credit is one facet of overall economic policy - increased availability of credi A + ay h-ave itt ~l e eflect or. proluctior, if other ecor.oruic polic4les Muilitate, ~A. ~ S. SiO ssv .LJ. . L. L& L~. SiL LUUL L.A.L. JLL 1. L. L LL JLIUL. jJS.L. L. ~ SUJ.L. La.A against farmers' using credit profitably. Where such a situation prevails, crei wil 4rl beom efeciv i -I-I 4 4f those -- - po4cies4are -Adif4eA.)t 1.52 Cfredit an' alternative programs of the ki.dindiae c-et - for- I . .'. ~ .J. U A..L cLU L,.LL L L.LV LL~ JS J. L.Lt= £S..4LLL .L&1U.LL;UL.U .L PU/ L-. scarce funds to achieve basically the same objective, i.e., to encourage fariL,ers to adopt new agricultural practices. lhe cho-ice of mear.s, or cobl-b= L1. C~ ~ LU L. ..I~ . LU.L U a. L A. L. A ±LL& L.L UL A.~ k U~J ~ I L, U A. nation of means, should depend upon careful analysis of the nature of the constraLints iL a spec1.ific sLtULt.L[L, a-s well as of tLie costs o' alter..ative programs and their political and administrative feasibility. The costs and the ability of the governmert to deliver services to a target group, espe- cially when those are small farmers, can differ markedly among programs. 1.53 For the very poor in rural areas, agricultural credit programs are of limited application. Many of the very poor in the countryside are landless laborers, or those who are too aged or too infirm to work. For these, credit will be of little direct benefit. But even those owning a - 30 - little land may either have no productive opportunities or require such small amounts of capital that grants would be more economical than the cost of col- lecting loans. 1.54 To help these underprivileged groups, programs other than credit will have to be adopted. Agrarian reform is important. Programs aimed at improving the productive opportunities of small farmers will also help. To date, most agricultural research has been directed at monocultural prac- tices and little has been done to improve methods of intercropping, typically employed by very small farmers. In addition, programs not directly connected with agriculture, such as health and educatior schemes, are also likely to improve farm productivity. But in the end many of these people will improve their lot only by leaving agriculture and finding employment in more productive sectors of the economy. - 31 - II. AGRICULTURAL CREDIT PROGRAMS AND POLICIES 2.1 The success of credit operations depends not only on the conditions under which farmers will use institutional credit to expand production but on the aaricultural credit agencies themselves. With regard to institutions, there are three key problems: the design of credit programs, the financial viAbility of the .redit agencv and the structure of the credit delivery mech- anism. In order to discuss these questions, however, some conceptual clarifi- at-inn are rpnirpA. 2.2 First, for purposes of xpnosn tionn thin naner divides farmers into two classes - large farmers and small farmers. There is, however, actually a continuum of farm sizes And the problems of the meduim-sjze farmers should not be ignored. Experience has shown that the most progressive farmers, in fact, the rean,l innov--tors, most ofter. come from, this c18s8. econ.A 1 n.nA quality is equally as important as land size. Three hectares of irrigated lowland may be far more productive than. 95 hectares of ar4A orA ntainous land. Third, the point was made in Part I that investment programs would succeed only if the fa -.er had profitabhle --ortu.ities in which to ir.vest, but these opportunities are of various kinds. To take advantage of some requires only add.AAz44- l womng1 , capital, hil ot-he: n--- --- e aA.1¶tn fixed investments. Because of differences in the size of holdings, the qual- ity of lan.d ar.d the nature of the- ineten poru,t , here is a subjs tar.= .5 L. ~.L .L U ~LU L.&& L LLUL J. EUL &.&A= .LV ~O LULl- L JjljlJL ~.t.LiL , U.S4L J. a tial range in the additional capital required by farmers. This amount cannot be preciLsely ider.tified -.-ith farm silze, certa-inly r.ot with the oversim=~ U#J A. JtLO L .L.LL %.LL .L U W.L LAi£LUA LA L~ L U.LL.LJ iI. L.L& Li 'I&L&. plified rubrics of small and large farmers. 2.3 For the design of credit programs, what matters is not so much tIh size UL JaLU L1VlU±AgL, UUL LAM scale arlU type ofL the faL=LULC Luan Lreu.LC- ments. For example, the costs of administration rise as the average size of loan ueclineS. DUL a small farmer sei.nLL to LLL&dLnc tl. cquisiuin ul d tubewell may require more funds than a large farmer borrowing in order to purchase fertilizer. Secondly, the availability of credit personnel and the level of their salaries greatly influences the type of credit program tlha can be economicaily administered. wziere saLar±es are low, LL iL, of course, possible to service smaller loans. In summary, on the rich alluvial .Lanus in certain areas or Asia, it may be economical LU thave creuLt prugrams serving farmers with much smaller landholdings than, say, in the mountains of Latin America, where, because of the poorer quality of land, the dirferent nature of the investments and the higher salaries of credit officials, the costs of credit service for farmers of similar size would be much higher. 1. Designing of Credit Programs Separation of Progrms for Different B---ao.-ers 92.4 Exprinc -with both Ban Grou pror --hA _ hol- U oO _U f-- when a single credit agency serves both large and small farmers, most of the fun.ds, if not most of the loans, go to the -arger famers. In part-, is is a political problem; at the local level, it is the larger farmers who have t-he olitica4r-l a.A .,o-14nC.no .4-,f11o 4-.,a.-. T~ _AA4f-4__- 1_.J_ the . pl.tical. at.d ocital 4nfsca luence with th-e credit ager,ts. h.*flt. L.. %.LL L. Ir. a4A.iLLJVL, .LILUe ing to the large farmers involves lower costs and lower risks. As will be diLscuassed" later, administrative costs on small loans to widely dispersedU farm- ers are expensive. For many reasons, institutions are often unable to charge Interest rates high ernough to cover the costs o' small 'amer lend.g. .L 1A.L L& I&~L UY L &LL L .OO O OUL.L A L_IL L C ± Luig. Because it is more difficult for small farmers to generate a marketable surp'Lus, thLe -risk. of1 dUefLauLt associLatedU wiLtlhL small 'iarmer lUndig Lb also greater. Lastly, many credit agencies require some form of security from Larriiers, and small fLarmers Lha-ve l'ttLe or no secur'ty to offer. Because the misuse of funds for consumption, high default rates and inability to cover costs are usually considered the hallmarks of a poorly administered credit program, managers of credit institutions attempt to hold down the value of small iarsuer loans 'L oLULd to -meet thie criLerLa Dy WhiCh they are judged. 2.J It seems advisable that the Bank snouid support separate credit programs for large and small farmers, in order to increase the probability that funds intended for small farmers actually reach small farmers in areas in which land holdings differ markedly by size. But, since credit institutions utilize scarce human and capital resources, it is often not economical to have completely separate and parallel organizations for small and large farmer lending. In such cases, separate programs should be created within a single institution, to a large extent utilizing the same facilities and personnel. However, the financial accounts should be kept separate, and many aspects of the programs will be different. Where there is not a clear distinction between small and large farms, it will be necessary to adopt other criteria such as income to sort out the two groups. Providing Credit for Particular Purposes 2.6 Although Bank Group agricultural credit projects have generally been oriented toward funding specified production activities (hence their classification into livestock, mechanization, general agriculture and other categories) the items covered by such programs are not usually exclusively of a single type. All components of a production package should be, and usually are, financed under such schemes. In some cases there have been unfortunate instances where this has not been so. As an example, in cash crop projects in Africa credit has been provided for inputs to the cash crop (mainly fertilizer) but not for the subsistence crop, though the cash crop can be grown on many farms only if land is released from subsistence crops by increased yields. In a number of agricultural credit projects, as in India, short-term credit for incremental working capital required as a result of permanent improvements and adoption of modern technologies have not always - 33 - been made available to farmers. This resulted from the lack of coordination between two separate cooperative systems, which provide either short- or long- term credit. Clearly, there is a need to think of production packages for the farm as an entity and to finance all complementary components of a new techno- loyV uackaee if the goal of rapid innovation is to be attained. Ftwther. there is a need for related long-term financing plans which project farmers' overall financial recuirements and recognize the necessity of sustained financial support for on-farm development and working capital. Terms and Conditions of Lending 2.7 In the past the Bank Group has financed credit programs lending to farmers for investments in fiwae capit I - hbl i ung this tvnp of lending to be most appropriate, since it is the lumpier investments with longer term pau-nffs that farmers have the mnat tronbhlp finanring from their own savings or from alternative sources such as money lenders and merchants. (The latter uQually provide onil relatively short-term credits.) MAny of the lonaer term investments in fixed capital are not suitable for small farmers because of size indivisibilities. However, the creation of farmer groups to share capit fl investments such as tubewells and pumps and further reduction ia the minimum scale of such 4-vestment such as the development of rotary till-rs and smaller irrigation systems can reduce this constraint. At the present time, however, ,uch -C of .n. t _11__ fa-4rmers 4,natnmaent 4i for sasoq nal 4n-,nat,nan1-o SL.tI I .JA LL.- -aULaJ..L & *0.5 -lA~ C -~ V tC - '.fl.* & A. i -. ' _..4l* C _ - to take advantage of new technology comprising seed, fertilizer, pesticides andU the lkIe. ~LLU LC .L±N . A . 0 iiuudy, LLLc JL U XLLU=L LWl UL.L %6L CUJ.. A. L.JUt J9UU.L.L_ 05i. L&.L1 UL LL UL b-%UX institutions is for short duration loans for working capital. Annex Table 11 sliows that ir. AfLri Lca a,.d LCdatin ZUIiUL CL 7 5 percent of thle instiftuti.onal loans are for two years or less and that in Asia the figure is 65 percent. Cer- taLinly, a similar propo r tion of small farmter creuLi; reeds will bue shLort-term. The Bank Group has, for the most part, followed a policy of financing a working capital revolving fund onl-y when it constitutes an essential part of in.-vestmernt under a project and meets two conditions: it is incremental start-up working capiLta'l andU such Lir arLl,c L [oL avaiL.LbLeU. fLUIU ULoteL oUUrces. LLM Ba.-,k's approach has been essentially pragmatic, however, and in recent years the pol'cy toward seasonal lending has been relaxed. One of the ''rst casss of such a change in policy was in Tanzania, where the Bank Group financed a substantial portion of a permanent working capital revolving IunU requireu by supply cooperatives to purchase and distribute fertilizers and pesticides to their members. In the future, the Bank Group's increased emphasis on small farmer projects will necessitate more such lending to credit institu- tions engaged in financing seasonal activities. Credit of this kind Ls most needed by farmers to produce a marketable surplus. - 34 - 2. 9 The small size of individual loans is one cause for the high admin- istrative costs in most small farm lending programs and these programs can ill-afford to fragment further small farmer credit needs into separate long- term and short-term components. Consequently, small farmer credit agextcies should be responsible for the delivery of both long-term and short-term credit, supplementing long-term financing with working capital when needed, as well as providing short-term loans for small farmers with only seasonal credit needs. Provision of ComDlementarv Services 2.10 As discussed previously (paras 1.39-1.54) a shortage of capital is often not the only constraint preventing farmers from adopting a new technology. Often the farmer lacks other imnortant ingredients - know]edge of terhnolovy, inputs such as fertilizer and seed, and, sometimes, even the ability to sell his crop. To mnke the red it nronram a sutiress the government must provide the complementary inputs that the market system or others do not provide, or flrAo,iAp bait poorly. UIRsu1 ::ally vao,s norinl7l 4 >PP1d non1provide these auxl-Yl r r~~~~~~~~~~---- - -S -- ----- -o- -- - -- r -O --- iary services, such as extension, sale of inputs and marketing. In other cases, hovpupr seeural sePna-ratP servirPes are delivperped inntlu in what- has -, ---_- -_ - _ ,…-_--… ---- - been called a package program. 2.11 The package approach is to be preferred in that it provides the farmer noft onl ry 4redt buht mll nf tho nr4 11 o,r vces hea requiresn4 eea n though some of the benefits of specialization are lost. Several of the small- holder projects which the 1ank. has financed in Afrlca areof the ntegrated variety: the Lilongwe and Shire Valley Projects in Malawi, tea projects in Kenxa, Tanzrania and Mauritius, t4h iJ-lm.o Pro4ect 4i E-hoan--, the la-.- Project in Senegal, and the cocoa and palm oil projects in the Ivory Coast. These projects are consiAered quite- ]---Asul Onteohe rd i eea i. *t~a. J -.J ... o atL IAfOS _ 4t L - I _ aL..AcO-t . -JS thei 'J ae LLS m ul , LJA 0 CV=ra 41 credit programs in Latin America in which the credit agency assumed the respon- s4ibilit +y for perform,in non=credit functions which, were not Airectly relatedI to the purposes of the loan, there was a deterioration in both the quality of lending andu in th rdtinttto'sfnnil oil A. LUJAL~ 1 Lni L1L%-Z LiJ U-L L. k1LML..LL.U#L.LLJLL Z0 LJ.Li411 4.U1 jJUO±L±U.LLL L..12 bi Lie ther a single Uaggency shoLUU±U Ub LrbpUoLn.LUl ior delivering bevera-L services or specialize in the provision of one is situation specific and depends upon the availability of traineu personnel and the degree or institutional development. In countries rich in manpower and institutions, there are some advantages in specalizaLoUn. In CuunLLUres Uomewnat poorer in DOth, it is possible to separate functions at the apex level but combine them at the local level where the availability of staff is thinnest and application most costly. In countries with the least developed administrations, it will be necessary to utilize whatever distribution channels exist. Sometimes this will mean credit officers providing extension or the reverse. In other cases, organiza- tions developed by private companies to ensure adequate supplies of such crops as tobacco, tea, cotton and groundnuts, can be used to provide the needed serv- ices. In yet other countries it will be an area or crop development organiza- tion, such as those being financed by the Bank in Africa. - 35 - 2.13 In summary, the scope of activity best suited to credit agencies will depend upon the specific circumstances related to the availability of trained personnel and the feasibility of using or creating separate administra- tions for providing the additional services required to stimulate farmer develop- ment. If a multi-service approach is adopted by the credit institution, the costs and income from various activities should be kept separate and the cre- dit agency should be reimbursed for the costs of services not related to the administration of the credit program. If a clear division of costs is not established and if ancillary services are not compensated for, it will not be possible to judge the financial success of the credit program, and losses incurred through the provision of ancillary services may lead unjustifiably to the abandonment of the entire program. 2.14 The most common ancillary service provided and financed by credit agencies is farm supervision. The degree of supervision employed varies widely, covering the range from the simple provision of basic crop informa- tion, a service roughly equivalent to extension, to the specification of in- puts, sometimes provided in kind to prevent the misuse of loans, to almost complete control of the farm operation by supervising officers. Supervision is designed to help the farmer but also to prevent loan funds from being mis- used to finarnce consumption, and to insure repayment. Improved supervision can encourage the effective use of credit. 2.15 But supervision to prevent the use of loans to finance consumption can be only partially successful. Farmers and particularly small farmers seldom divide their operations between production and consumption. Farm life is integrated and much of their own production is used for household consump- tion. Owing to the inherent fungibility of credit, supervision can never completely eliminate increases in consumption subsequent to the receipt of loimns, even when credit is provided in kind. Given that supervision is costly, it must be subiected to careful evaluation and planning in terms of number, purpose and scope of visits by the supervisory officer, to be certain that the benefits iustify the costs. In the main though, expanded super- vision appears to be warranted as a means of reducing defaults. - 36 - 2. Financial Viability of Credit Institutions 2.16 Throughout the period of Bank Group involvement with agricultural credit, a maior concern has been to strengthen the credit institutions within the borrowing countries, and particularly to ensure their financial viability. The main reason underlying this orientation is that institutions without financial viability, if they survive at all, are dependent upon annual annrooriations from zovernment to helD cover costs. and are therefore susceDt- ible to political influence. In fact, without substantial subsidies, few exiqting credit institutions would have been able to survive since, for most of them, costs exceed revenues and inflation plus defaults have eroded their cafli fal stru_rtu,re. LTn_Ti Delinoiunrcie - Oierdues and iDefaultsi- ?.17 Fa-llure, nf fnrmers t-n repay thpir tde-hf- on timp- or even tn rpnay them at all., is a serious problem in most agricultural credit institutions. The data on delinquency arnd defaiultt hv o-nuntry and program are shown in Annex Table 12. The table presents two measures of arrears where available, unpaid loans as a perrontage of thp tntai nortfolin and as a npr.pntage nf pavments due during the year from both loans coming due and those overdue. This latter percentage is called the arrears rateo The figures cited should not be used to make invidious comparisons among countries or programs because the fi-gures reflect wd varyiation in definition and in ruuallit-y of lnfnrmTat-ion. Nevertheless, the import of the data is clear - in most programs, delinquency rates are ver- high, frequently as much as 50 percent of amounts due. Some agencies are thiought to have even higher rates of arrears than reported in these tables, bDut theose aro e con.cealed, primarily through the refi4ncir.g of unpaid debts. 2.18 Data on actual defaults are very scanty. Experience shows, however, tlat except for a few counUries, r--a- of ------- pt s of ares is usually possible over a number of years, and on Bank-financed projects 'losses result'ing fror, defaults lave seldom exceeded fivre peren of loans i.UL tU i.t dl Uta La tId I ~ ULL U LCLtZU S .LVC I..LkCLLU ~~ ~LV.2LC outstanding. Nevertheless, loan delinquency is a serious problem for most dagrLicuLUl dcr Ce-ui L .L1L£ Lis LUtLoLIIb U'DUec Xu * L reUsu Lb J11 WdasL UL IUcIIapower, higher cost of administration and slow turnover of resources. Projects fi1nanced by th'e Bank- h'ave exper'lencedu seriLous collection problem-, in recent years in Colombia, Pakistan, Senegal, Tanzania and certain of the states of Iulia. 2.l9 There are three general reasons for overaues. Tle 'i.Erst stems from the farmer's failure to use borrowed funds for productive purposes. Second, overdues may result from adverse outcomes from the investment rather than from any failure to apply the loan proceeds as expected. Causes include bad harvests, natural disasters of various kinds, and changes in economic conditions which cause farm prices to drop. Some loans have been made on the basis of unrealistically favorable assumptions about the probabie results; at other - 37 - times, the terms of the loans were ill-suited for the purpose for which they were issued, e.g., short-term loan for medium-term activities. Although much of this can be prevented by improved supervision planning and a better appre- ciation of the real developmental potential in specific situations, along with considerations of the borrower's repayment capacity, many overdues will continue to occur for reasons that cannot be foreseen at the time. 2.20 The third reason for delinquency or default is not related to an inability but rather to a refusal to pay. Having the funds to repay a loan is, of course, not an absolute matter. Some funds are usually available and farmers have to establish priorities for their use. Apparently repaying public sector credits is accorded low priority by many farmers. In some cases, farmers have the impression that credit is a gift made to ensure their loyalty and future support. Governments sometimes do little to change this attitude, and may even encourage it in times of political uncertainty. Low interest rates may encourage delinquency. too, especially if new credit has to be obtained at a higher rate. 2.21 The farmer's lack of enthusiasm toward repayment is aggravated by an observed general unwillingness of governments, through their credit institutions, to impose sanctions on those whose debts are overdue. If land is pledged as collateral. government credit institutions rarely foreclose. Denial of new loans is the usual penalty for failure to repay. This is often a weak sanction., especiallv for short-term credit- since if the size of a recurring loan levels off, the farmer has less incentive to repay. This is reflected in a decline in repnavment percentages as programs mature. Lack of proper records and of an effective collection procedure also contribute to poor repnavment nprformanre 2.22 Cases have arisen where non-repayment has heen tho result of a concentrated effort to cheat the credit institution, sometimes encouraged by landowners and moneylenders fearful of th rnmnpetit-inn At times, it stems from corruption within the credit institution itself, when officials are more inte-resrted in brihes from those rveceir4ngY lo%ana thon in nthe diffivilt and personally less remunerative, task of recovering the overdues. 2.23 Failiure to repay is common to'large and small farmers alike. Small farmers would appear to be more prone to dellinquency stemm.ing from the first two causes mentioned above. 'They are more likely to use borrowed funds for consumption purposes and in poor crop years they are less able to generate the marketable surplus needed to repay their loans. On the other hand, in several count-rieo for exrnnlp Baoniei1A47, Boivia, i Com.b o Ricfa- aind Etthiopia, there is evidence that larger farmers actually have poorer repayment reonrd.a In mnry of these cases, it appears that large farmer delinquencies are deliberate. Large farmers use their political power to protect themselves a galnst the penalties for del4nquencies ATheir overdues alo occur w agrarian reform measures are expected or already in effect, and they stop rep-aying past loans in tbe h-ope that a debt adjustm,en.t or moratorium on rep'.~ ayment, Vaa C *'.'~ * .. VL. w ilU be f hJ0oaing. repayment will be forthcoming. - 38 - 2.24 From the overall economic point of view, default is a transfer payment to the defaulting farmers. But it is one of the least desirable or e.uiitable forms of transfer: it destroys the financial viability of the credit institution: and farmers who know they will not be required to repay are more likely to use the borrowed funds for consumption purposes. From a social viewpoint, this is one of the most costly aspects of the default problem. 2.25 Reducing the levels of delinquency and default is the most important issue in seeking to make public sector credit programs financially viable. Traditionally, credit agencies have required that borrowers pledge some collateral ! usually land, to secure their loans. Although this practice is feasible when lending to large farmers, small farmers often lack certified titleso o their land and tenants have no title at all. Moreover, it is frequently difficult and expensive for the small farmer to obtain the appropriate legal instruments. For its part, the Bank has consistently enmphasized that the productive capacity of his holding should substitute for security as the essential criterion in loan decisions. 2.26 In many cases, it is too costvly to foreclose on assets pledged by small farmers, making security instruments poor protection against default. Lending only to those wiat-h in,estment nnnortunities sufficient to produce a significant marketable surplus is perhaps the best way to reduce the level of default. In such prograTs as the Pupbla proiect in Mexico; INCORA in Colombia, and ACAR in Brazil, where credit was followed by a large increase in output, the problems of default and delinquency have been noticeably reduced. Another practice which has reduced delinquencies is to coordinate repayment wlth markleting of crops wrhich m,ust be centrally processed., e=.; tobacco. cotton, cocoa, tea and coffee. The use of chattel mortgages and liens on crop production has also been effective as a low-cost method of protecting against default. 2.27 Following a bad harvest, credit agencies frequently adjust repayment terms , eitihler through renewals or postponements of maturity dates. However; most of thie available data indicate that once a loan is in arrears, collection is both dih Licu'Lt and cost'Ly. Thls suggests that in areas where output is higlhly variable, it might be possible to employ contracts whereby the credit agency would be paid a percentage of the farmers' output rather than a fixed amount. Such sharecropping arrangements are quite common for land rentals. Crop insurance is a possible way to protect both the borrower and the credit agency against the vagaries of nature. Both approaches involve difficulties of administration and ia the case of crop insurance may prove costly. 2.28 If the incidence of delinquency arid defau'lt associated with poor harvests, an(d from failure to use borrowed funds for production, could be reduced, credit agencies couid afford to dea±l more strictly withl thLe remaining source of overdues, i.e., farmers who have the funds but refuse to repay. The onlus attached to attempts by puDlic credit agencies to take court action would be greatly reduced if such efforts were confined to the deliberate ,lefaulters. This becomes largely a matter of attitude and political will. A decision on the part of the Bank Group to withhold funding where that will is lacking might strengthen the government's hand. - 39- Costs of Agiricultural Credit 2.29 Credit programs are costly to operate. The administrative costs of agricultural credit institutions vary considerably, as evidenced in Annex Table 13. In part, the difference in reported costs reflects a difference in concept and in the responsibilities of the institutions involved. The costs of supervision and other ancillary services vary, and eliminating these items to achieve comparability among agencies has not been possible. Further, in some cases, the cost figures cited encompass the entire credit delivery mechanism, while in others the costs refer only to a single element in a many-linked chain of credit delivery. In addition, cost accounting procedures, especially as regards reserves and write-downs of defaulted loans, vary.widely. For the group of institutions shown, the median figure for administrative costs as a percentage of the total portfolio would be around five percent. 2.30 The administrative costs of agricultural credit institutions tend to be high relative to most other types of lending institutions. Because borrowers in rural areas are widely dispersed, credit distribution is more costly than in urban areas. Also; collection costs would be higher as a result of the high level of overdues inherent in agricultural credit. 2.31 Administrative costs, of course, depend upon the size of loans and their duration. For an efficient institution making medium-term and long- term loans to large farmers, the costs of administration are about three percent of total portfolio. 1/ The costs would rise if the financial institu- tion provided ancillary services or had to mobilize funds through deposits. 2.32 Although a general study of costs in credit programs supported by the Bank Group has yet to be undertaken, a recent analysis made of adminis- trative costs of Indian credit agencies funded by Bank and IDA loans may serve as an example. At the level of the Land Development Banks, the adminis- trative costs were three percent of the total portfolio. Adding the costs of the apex bank raises administrative costs to four percent on outstanding credits. However, in India a portion of the total costs of credit adminis- tration are borne by the Government Department of Cooperatives, and if these costs were included, total administrativo, costs would be higher. In addition, in India, the Land Development Banks handle only medium-term and long-term loans going principally to medium and large farmers, provide no extension or supervisory services, and benefit from a salary scale that is low, all of which keep costs down. 1/ This is somewhat higher than the administrative costs of DFCs, which lend long-term to large corporate customers, and somewhat less than the costs of efficient commercial banks in developing countries that lend to larRe urban corporate customers at short- and medium-term but have to incur the substantial costs associated with savings mobilization through deposit issue. - 40 - ^.33 The costs of adm.nistrato r4se as- th-e size of loans fals,asth - . ~I. IteL U L L ~u LLLJ L .L U LLLL I. ~ J- C LAL ~L .C JL ±JaLL LE.LJL Q 1C L.L1 duration of loans shortens, and as the amount of ancillary services provided j-------- .I- -lonosstrd to be greater in LmoreV-L w=eaalth coILwe --wh .L LU L *,13U tUPiUL Z5LCUZ LZ~ L=LLU LU UC , CcLC .L U WdL Ltl1 CUL-JuLL tries 'Lnhicb credlit officers are paid higher salaries. For an efficient small farmer credit institution providingL a g ix of short-term and long-term loans and itself bearing all of the costs associated with credit delivery, total administrative costs, excluding extension and otner ancillary services would be between seven and 10 percent of total portfolio. 1/ The actual cost would depend upon the size and type of loan and tne saiary scaie. Tnere is a trade- off between administrative costs and delinquencies and defaults. The more carefully the institution scrutinizes the applicants, supervises the ioans and p'irsues delinquents, the lower the delinquency and default rates but the h:Igher the administrative costs. However, no amount of appraisal or loan supervision can reduce them to acceptable levels so long as they arise from political misuse of the credit institution and tacit condoning or failure to repay. With sufficient political will, it should be possible to reduce complete defaults to five percent or less in a well-run credit institution. 2.34 There are also the costs of the capital used by the credit agency. Some public credit institutions receive government loans on which they are required to pay little interest, others operate on government-provided equity funds. wnatever the financial charge for capital, however, there is an opportunity cost of using funds for agricultural credit rather than for some alternative program. There is a substantial literature on the opportunity costs of capital in developing countries. These estimates are seldom less than eight percent in real terms, approximately the level required to mobilize savings effectively. 2.35 In summary, the foregone opportunity costs of using funds for agri- cultural credit are at least eight percent per year in real terms. Adminis- trative costs for institutions making i.ong-term loans to large farmers may be as low as three percent per year, while for short-term loans for small farmers, these costs will average at least eight percent per year. "Normal" defaults can be expected to add another four percent to costs. Excluding supervision and ancillary services, total costs in real terms for an efficient institution would range between 15 and 20 percent, depending upon the nature of the operation and the average size of loans. Interest Rates 2.36 Credit programs in the past were aimed at protecting small farmers from exploitation by money lenders. Consistent with this goal, interest rates were set quite low. It would be fair to say that these programs were not expected to be self-financing; governments, at least implicitly, recognized 1/ In developed countries, consumer credit institutions which lend relatively small amounts primarily at short- and medium-term have administrative costs of between six and 12 percent of total portfolio, depending on the term, collateral, and credit risk involved. Default experience averages around two percent. - 41 - that they contained a large element of subsidy. Although the original program objectives have changed over time, the interest rates charged have remained relatively low in nominal terms, and in some countries with substantial rates of inflation they have become negative in real terms. Institutional interest rates are low in relationship to those charged by commercial lenders, low relative to the supply and demand for funds, and low in comparison with agency costs. 2.37 The nominal rates charged farmers by institutions (as shown in Annex Table 9) fall between five and 30 percent per year; half the institutions c.harge between nine and 12 percent, with the remainder roughly equally divided between those charging more and those charging less. Interest rates eorrected for inflation range from minus 16 nercent to plus 16 Dercent. The nominal rate charged by institutions averages about 10 percent and the real rate above npercent= 2.38 From the farmers' noint of view, institutional credits involved other costs in addition to the rates cited above. Detailed studies of the true costs to the farmer of official agencv credit in Banaladpsh, Brazil and the Middle East found that the combination of application fees, travel and "entertainTnt" costs, and working days lost in arran£ing lons g'reat1v diminished the attractiveness of public credit as compared with private borrowing. Inform.atinn from other coiintries suiggests that significant "informal" charges on public credit are not confined to these few cases. Nevertheless, eve ,n 1 1 ny.7 i na fory. t-hese hi ei naot- a 1mne-si i4nstitutionnal c-redi4t- ini mnczt countries is notably cheaper than most loans from commercial lenders. 2.39 The level of interest rates on lo&'ns to sub-borrowers and the spreads available t1o credit, 4nstitutions are recu--,--t point- of disusson4e- er ~~h¶71l 1 0h 1 n ~~~~~~~~~~~~~~~~n ,,~~~~~r,.A 4 4- 4 not- 4 4-,. 4-4 no.0 o ra -rnnnrrrlr, 1- nn4r. 4-a nE' .44 en. .oe4 n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. k0 4-r.,IO-- n-o.-- -- - the Bank Group and many borrowers. It could be argued on economic grounds that #.11ese ;n LX _ rates S .UI A. rA- _ Ac 1-U-e f Au ct.tIO. - capi t. A A t -*fe14_n m rests on an assumption that even though capital/credit markets are imperfect, thie interest rate influences resource allocation, savings mobilization, financial viability of the credit institution, and equity. These points are considered 4n +-urn. no . 1O A 1 1 e * 4 AnO..A 4 _ . 4e "I Ank 4 - v 4 A 4 t. . i 4 A- U ,7 4 . wi_ L V . 9U LCC vL_UU L.C nt JL L A-US- A- L L A-C .L --e, *t X- - _fi U.LSJLDO v-WA.. E ,& production orientation, financing investments with a high expected rate of return. 'ILA productive opportunities exist and capi.l- -i6 crc, - lo- w rates should not be necessary to stimulate investment. On the other hand, if Larr,.ers can afford to pay or.Ly low rates bUecause thUeL in.vestrLILItents opren to them have low yields, then it is questionable whether these investments constituteL theI bLLest use of scarce resOurCcs. LoUW intLCLest rates leadU to higher capital-output ratios on the farm. In countries with surplus labor, ttiLLs 'Ls a questionable poUliCcy. - 42 - 2.41 Savings Mobilization. Agricultural credit agencies have attempted to mobilize rural savings in the form of both deposits and equity participa- tions. Those countries that have succeeded in mobilizing voluntary savings in the countryside, such as Japan, where deposits in cooperatives amount to 84 percent of working capital, and Taiwan, where the comparable figure is 76 percent, have paid high rates for depositi. The savings mobilization effort in other countries without high deposit rates, such as India, Peru and Senegal, have been much less successful. 2.42 Some institutions offer equity participation as another means of mobilizing capital. In many countries (Ecuador, Thailand and Tunisia to name a few) farmers are required to purchase shares in the cooperative in propor- tion to the size of loan. Farmers regard such purchases as compensating balances whose effect is to raise the rate of interest. The difficulty members have in redeeming shares, the absence of dividends, and the poor operating record of many of the cooperatives militate against persuading farmers to voluntarily purchase shares in most countries. 2.43 Financial Viability. Obviously, financial institutions that charge rates that are insufficient to cover costs cannot be financially self-sufficient, and if they must rely upon government transfers to survive, they become more susceptible to political influence. Where governments are unwilling to provide adequate transfers, or if politicization of the institution results in high default rates, the credit agency runs the risk of paralysis and ultimately of reorganization or closure. Financial institutions represent an investment in organization and human capital that few developing countries can afford to lose. 2.44 Equity Considerations. When the charges on institutional loans arc set unduly low, large farmers because of their disproportionate political power, almost always garner a disproportionate share. In fact, if cheap credit is available in excess of investment needs, large farmers may utilize instittitional loans to purchase land, thereby exacerbating the land tenure problem. Evidence from Colombia and Ethiopia shows that this is a serious problem. Unless rates are high enough even for small farmers, the replenish- ment of resources of the financing institutions will not take place and fewer farmers will have access to credit. '.45 There are, however, a number of major considerations which in nrartice reaiiire substantial departures from using an interest rate that covers the real cost of lending. These include: (a) Comparative intersectoral interest rates. In practice real rates throughout most economies are likely ton h well below the rate that would cover the real cost of lending in agriculture (especially to small farmers) Forcing Un intereQt rates foir agrficult-uvra nlone cn lead to an uneconomic diversion of resources and considerable leak- aIe bertweOn sectors The rl robem for nar icult sure cannt 0 beASepra from that of the economy at large and a solution rests on restructur- ing lending rates -as a 4ho in t-he cont-xt- f nat-ional pn14i on interest rates. - 43 - (b) The leuvl of intprpet rates and the nsyvhological impact on borrowers, If interest rates are to be raised substantially then this could wpll havp a dplptprintii impnact on horrowers Sharn rises in intpr- est rates may well deter small scale farmers or those farmers moving into th:e cash econ.om«y fromi hnrroing the resources needed to sustain. production of a marketable surplus. Increases in interest rates would have to take plc gradually to avoid discouraging this class of borrowers. Interest rates would probably have to remain low over f fn4lrl long period of t4me Tc ow~ levels of interest rates an-l d4scr4_4-#-4__ -^ar.t- -4rculture. The terms of trade are often weighted against agriculture and low 4n er es t e e care a_v oA*evL A as _ ,d m _o a ns At ofse *enn h AAi e d_.c rm;4a- tion arising from adverse terms of trade between agriculture and industry or agriculture anA S ___. th e- rest - the ecor,o.,y -e mlay well be merit in using interest rates for this purpose. However, by the same toklen it muay well b-e that modifyi.r.g th ntrstrtet oaL , JJ'U LL . L 4.ny W .LJ Li L.LL"tL &ILU%±ALA .LiLLLr, "t A.LLA L.L n . ai. offset the terms of trade may also add another distortion to the ec UnU ioy. a L1 LtLLs event ±L XIray ue prLefLrabU Lu go LUto Lthe iLLar L U the matter, where possible, and change those policies that do discri,minate again>st agricuLture. ±Llese raight incluue polucies such as tariff policy that leads to high cost domestic production oIdma.Luiactured goods or to policies that penalize agricultural exports. (d) Low interest rates help to fulfill social objectives. Subsidized interest rates - especially to small farmers - can heA.p redistribute income in favour of the low income groups in the agricultural sector. Given the extreme inequalities in income distribution that do exist there may well be considerable merit in using subsidized interest rates for this purpose. However, there are also arguments against the use of subsidized or differential interest rates for social purposes. (i) As has been pointed out above, lower interest rates lead to higher capital output.ratios on the farm. Low interest rates may be a questionable policy when there is surplus labor and may militate against the expansion of employment. (ii) I.xperience indicates that subsidized or differential interest rates are open to considerable abuse; frequently there is leakage and larger farmers obtain loans at subsidized rates. Subsidized interest rates also provide opportunities for corruption and political manipulation. It is difficult to ensure that loans at low interest rates reach the groups they are intended to assist. It is not uncommon to find that subsidized loans have been diverted to non- agricultural use or used in a way that contributes little to the raising of output by small farmers. - 44 - (iii) If it costs 15 to 20 percent to lend money and the sums ch arged for !-endin.g i half If this amont (as they freuntly are) then each loan made by an institution will contribute to a depletion of it. s resourcei . T,I,4 S in turn will Al-i4-iis the 'reflow' of funds into the system for institutional 'lendi-ng. T.h e lending ability of L.h'e institution will' I- .L LU i1,* .L.±IUL5 a,A4 LL S.L1 Ll i U±JL W. L Lie weakened and the process will also weaken the financial 1.JJo the lnsttutilon. As10 is pUoi1tedU oUL aboUVe, reliance on budgetary support to maintain financial viability U1imiir.nis'es the institution's independence andU can rukae it susceptible to pressures which may reduce its effectiveness. (iv) Subsidizing interest rates may not be an effective means of redistribuLting lncomtie. WIIdLevCe ±SiLLULutlondiL credit is avail- able will only reach a minority of small farmers, It is questionable whether income distribution can be improved substantially by further subsidizing those small farmers already benefitting from access to institutional credit when this reduces the resources available to the lending institution to provide similar access to other small farmers. in addition, subsidized interest rates would not help the large group of landless laborers or the smali noiders whose holdings are so small that they cannot qualify to receive credit. Redistribu- tion of income can often be implemented more effectively by means other than subsidizing interest rates for small farmers. (v) Subsidised interest rates have also been advocated as a means of encouraging technological change in agriculture. Low cost capital may well induce farmers to adopt changed methods of production but subsidies to encourage change are usually most effective when they are linked to a particular technological change. Thus when there is interest in encouraging the use of an input such as fertilizer it is preferable to subsidize the fertilizer rather than the cost of credit per se. Further- more, subsidising an input such as fertilizer has the advantage that it can be modified as the input gains acceptability and so can he varied, depending on the rate of acceptance of the new input. Subsidized interest rates, on the other hand, have a pervasive effect and so cannot be varied according to the rate of acceptance of a given input or change. (vi) One sulbsidy that might be effective would be a subsidy to cover the added administrative costs of lending to small farmers. Such a subsidy could be used to equalize interest rates so that all loans of equal duration would carry the same charge whether for the use of large or small farmers. A unified interest rate would avoid the leakages and opportunities for corrtuption that might stem from differential interest rates; in addition, a uniform interest rate would help foster optional resource allocation, especially of the rate reflecting the true cost of capital. - 45 - 9JLAA Thp disctussion above indicates very clearly that there is no simnle or unique answer to the question of what constitutes an appropriate interest ratp for agrirtulture; esneciallv for small farmers. There are many factors that impinge on decisions regarding the establishment of an interest rate that suits thp npttern and needs of agrlrulture in a given area and also fulfills the more general objectives of national economic policy. 2.47 It follows from the above that there is no unique answer to the quesioanr of~ wahat might be an appropriate interest rate on a Banki financed agricultural credit project. Many considerations will have to be brought to bear. First, there are obvious difficulties in trying to'obtain different interest rates on Bank projects than those charged by the same or competing 4institut- aons on aeh er ebutsiisr pro4jct 1s - T4hA 4s e so iA 4 t n 1n thnc cs regions where other external lenders have accepted the principle that there shouldA be he1vil 4A4 sus zed Ainterest rates. Cl arl 4y, .- a., -h- -situatons the Bank should attempt to work with the other external donors and the borrow- lng gove.,,Lnent a.^.d nstitution to develop a ---..o.a policy that 4s consistent with the best interests of the national and agricultural economy. 2.48 In general, the Bank should work toward the objective of institutions ilenUding at positive interest rates thLlat refLect tle costs oL' capital ard of providing the capital. Such an objective should be seen as a long term goal. An intermluediate objective miLgh't be for the ir.terest rate to bue at a 'leve'l that at least covers.the opportunity cost of capital. Where subsidies are used tIese should, in general, be lim.Lited to cases -where they can be clearly justified and are likely to be effective in view of the pattern of farming in the project area. 2.49 There is a special problem regarding lending in an inflationary situation. In recent years, the Bank has made loans, particularly in several Latin American countries, where the rate of inflation is both high and variable. On several occasions the Bank has refused to make new loans until inflation was brought under a semblance of control. Where it did lend, the Bank insisted either on rates sufficiently high to cover expected price increases or on the indexing of t'ne principal of the loan and/or the interest charges, a procedure that has now been used by the Bank in nine countries. 2.50 Unless indexing is applied throughout the economy, it disrupts the allocation of resources if pursued for agricultural lending but not for other kinds. There is an adverse equity impact if only farmers, or worse still only small farmers, have to pay indexed rates while others in the economy do not. Further study of this inflation problem as it relates to credit is recommended, but this is a problem relating to all lending, not agricuiture alone. 2.51 Another difficulty is the choice of an appropriate price index to which to tie interest rates. Various indexes have been used, including the price of livestock (for livestock projects), alternative consumer price indexes, and the exchange rate, but all have been found to have deficiencies. - 46- If indexing is to be used in future, detailed studies of alternative measures are recommended. In some cases a catch clause allowing for an increase in interest on, say, notice of three months, may be an effective alternative approach. However, when the inflation rate is variable, inflationary expectations differ and high nominal rates, particularly when set only on certain loans (say institutional loans in agriculture), may only deter borrowers rather than promote development. 2.52 Finally, it must be borne in mind that credit is only one element in the package of inputs and services necessary to raise the productivity of small farmers. Frequently it may not be the most important element. Thus the attitude of the Bank towards the interest rate issue will have to be influenced by the extent to which the project overall is achieving the object- ive of raising the productivity of small farmers and achieving a satisfactory rate of return. The significance of the interest rate must be seen in the perspective of the many faceted requirements to raise the incomes of the mass of potentially viable small scale agricultural producers. III. SYSTEMIS FOR DELIVERING AGRICULTURAL CREDIT 1. Policy-Making and Administrative Issues 3.1 Decisions relating to credit program implementation can be cate- gorized as: strategic, tactical and day-to-day. The first category covers tihe broad priority questions: budgetary allocation to agriculture against otlher sectors and more specifically to credit against other developmental tools. This process covers borrowing from foreign organizations, such as the Bank Group, and setting key credit policy parameters, such as the interest rate. Because government credit can cover only a limited number of farmers, strategic considerations also include the decision on the characteristics, broadly defined, of those to be financed, which may represent a choice among regions, among income classes or among crops. In most governments these decisions would be made at the ministerial level. 3.2 Normally the tactical problems would be handled by the organization responsible for administering agricultural finance. These functions include: the detailed design and organization of the credit program; the choice, train- ing and supervision of staff; the handling and accounting of funds; and the coordination of activities with agencies responsible for providing ancillary services. In design, credit institutions are usually pyramidal: at the top there is an apex financial institution, then several layers of bureaucracy, sometimes within one institution and sometimes organized into separate institutions, and at the bottom is the widely dispersed organization which actually delivers credit to farmers. This bqttom layer is responsible for - 47 - day-to-day operations, that is, adapting the credit program to local condri t-4inns, -.-chsng 4indliv4 ,idual borr,r..-er nor organi zng7 lncya1l crpdi t gFrOupS, coordinating activities with other agencies at the local level and making, sune.rnicing and collecting loans. 3.3 This is a complex process and problem- can arise at any stage in the credit delivery system. However, there seem to be a set of issues, which, while not found ir. all credit organizations, amm. -n to mar.. Th rst of these is politicization of the credit delivery mechanism. This can hardly be avoided 4n one fo.m., or another wi.h the L stribution of an ir,put s.uchl. as credit which is in short supply and is channeled through a government bureau- cracy. B ecause Large fa..Uers hV ave greater political influence, th ey 'have usually been able to garner most of the government-provided credit. 3.4 In many countries, politics has invaded the tactical and even the operational 'level of credLit dUel'verLy. The choice of diJretor SLJL Lfo i t LhrUedL institution, and sometimes even of the staff, may be made on grounds of pouitical 'uva'Lt'Les rat'ler tnan quaificationfs. AtUU LU SUlUe :UUb pUL.L.LCb spills over into the choice of individual loan recipients and the degree to which sanctions against defaulters are enforced. Whnile the strategy of credit cannot be completely divorced from politics, it is obviously desirable to insulate actual operations from political interference insofar as possible. 3.5 The second problem common to many creait programs is over-cenLraliza- tion with subsequent cumbersome and inflexible procedures for procuring loans. An organization delivering credit to farmers, especially wnere the loans are small and many in number, has to be widely dispersed. Furthermore, at the local level the staff is usually thin and relatively poorly trained. To control such an operation is difficult, and the procedure typically adopted is to centralize the decision-making processes. Loan decisions often must be passed on to higher levels in the bureaucracy. The result is a credit program costly to administer, beset by paper work and red tape and not well adapted to local conditions: Illiterate farmers are subject to pre-audits and post-audits. The actual delivery of credit is slowed; few agencies can process loan appli- cations in less than 60 days and many take as many as 90 days to provide funds after receiving a request. For long-term loans for investment purposes, this time element is not so important. However, for small farmers who do not plan their credit requirements in advance, such delays in obtaining seasonal credit often mean that the farmer gets the funds only after the need for credit has passed. 3.6 Studies of farmer attitudes have shown that farmers often feel so negatively about the red tape and delays inherent in centralized credit programs that they prefer to borrow from the money lenders who can operate more quickly and flexibly, even though charging higher interest rates. Farmer alienation is increased when, because of inflexibility and the lack of a proper feedback mechanism, the credit program is poorly adapted to local requirements. The problem of poor staffing at the local level is a real one, but over--centralization is clearly not the solution. Institution building and staff training are called for. The use of farmer groups and cooperatives can also help reduce the problem of over-centralization. There is great scope for innovation and modification to improve the procedures whereby small farmers can borrow. 2. Channels of Agricultural Finance 3.7 As previously noted, non-institutional sources of credit provide most of the credit used by farmers in developing countries. On the institu- tional side the main types are agricultural or developmental banks, usually established by Government and partially or more frequently fully Government- owned; cooperatives or farmers' associations, often government-organized; and private or nationalized commercial banks. Many Central Banks are involved in various ways and to different extents in the delivery of agricultural credit, primarily as a rediscount facility but also in developing and enforcine lending policies and sectoral allocation of resources. In a few cases, such as in India; thp Central Bank has nlaved a considerable role as nromoter in the development of agricultural credit through research, guidance, supervision and nef -qbh1J.-hmPnt and ffnancing of acririiltiral rredit agenrieQs - ~ ~ ~ ~ ~ - -- -- - -- --- 3R The. funds nroivideid by the Bank GCrnou for agicltural crtrdit-r usuallv loaned or passed on initially to an apex financial institut;ion which then re1tlend eait-her directlv t-hronigh its own network nf hran-hcha es r threh independent intermediaries to the ultimate borrowers. The Bank Group funds hnve, therPforP hben rcurrpntlv handled at the ton level hb (Government Agricultural or Development Banks and Central Banks, or by special agricultural refin-anc f-nds usal loselyu assctiatedA wifth rCentral Banks. The Bank experience has been that Central Banks are well suited to serve as conduits for credit agencies and, where the objective of the Bank is to lend to fa_ers through many institutions, principally commercial banks, the use of the Central Bank for refinancing sa o ate. owever, C aBt- l Banks shou n 1o e too closely associated with the administration of agricultural credit programs, and espct ia11U 4in making 1oanso o ul ti.*te borrowers. This WolIA likely divert them from their primary responsibilities for financial and monetary policy. Thlere are also cssweethe borrow4r.g Goverm=.ent onlnsorpse n tLhe I. Iit L lL t L-JLi LIC WLIC. L&IC UJ.L L'.WJ.LLL '3VC:LLWICII L VL-JIL± LLLLU~ VL FLiii LLiC Bank Group funds to a Government agency or a special entity responsible for carrying out a projec; which relends the funds to the participantS in the projects. The use of such channels and of Central Banks as well should not obscure the need for building specialized agricultural credit institutions. 3.9 The volume of Bank Group funIds passing through t'he various channels which deliver credit directly to farmers is summarized in Annex Table 14. Commercial institutions, primarily banks, were the most important final channel utilized by the Bank Group during the period FY 1969-73, accounting for roughly 38 percent of agricuiturai ioans. The buik of these funds were for livestock projects in Latin America. Agricultural banks, once the most important channel, have become less important in the most recent period but still dis- tribute to farmers 25 percent of Bank loans. In recent years cooperatives have emerged as an important final channel; lending through cooperatives is concentrated in India but also has been used by the Bank in Niger, Korea and Tanzania. The channel described in the Annex as "Project Authority, Ministry or Special Entity" consists of the agencies or administrations, primarily government organizations, mainly in charge of regional integrated development, land settlement or development of a particular crop. The last type of channel, Development Banks, represents institutions which in addition to the financing of other sectors, lend to agriculture, mostly commercial agriculture. .3. Agricultural Credit Agencies Aricultural and Development Banks 3.10 In many countries, governments have establislhed these specialized credit fnsti4tutions mstlyv state-.ener or wit-h st-atP maijrity participationn for lending to agriculture. The Bank Group has lent to farmers through these institut-ieons in cuntrie scih as Afghanistan, Jordan, Kenya, Pakistan, Peril and Bolivia. These institutions lend primarily to large farmers, although their original purpose was often intended more for small farmers. Aaricultural banks are operated through highly centralized bureaucratic structures and thizs tends to nmake t-hem11 il-suited for lending to lag -m.br of 'highily dispersed small farms. As noted in the prior section, excessive centralization when dealing with, Smll -ar,mr ofte 4eut in -nrae adinstatv costs, an inability to adjust programs to local conditions, and political influence. 3.11 From, experiLence, there seemis to be definite advn ; in- spcil5e -J .I I£ I.UI~ CA~L .~tL C, ItC CLI LI. Ue ueJ. n Li LC 1'V MLI L04.,U Ii L4O£.C agricultural banks extending short-, medium-, and long-term credit over mUlti- scoa A eve_l1opmen _t _ __an|k_ 1 ,Avs --- 4-A1 __A _ 4 __ vA - - _)_A LA_- t i:>C::U_ LU L. Cii UCVC UJlJMILCIL L ULiitsO W LL.Ltk L C .LIL .L..LL=U L.U la. C L "'- II 0 ' LZL S.iiif agriculture, and to confine lending principally to long-term credit. Develop- meri,t bUankr.s are a'Lso centrally operatedU andU thus, 'U'lik-Ne agricultural bLanklIs , ill-suited to carry out small farmers' credit programs. Both types of nsLstitutions can, huwever, LendU to groups or cooperatiLves to mae their credt accessible to small farmers. Farmer Groups and Cooperatives 3.12 To overcome the problemLs associated with excessive centralization, many developing countrLes have lookea to tne estabDlshnUment of faLrmler groups and farmer cooperatives as the solution to the problem of small farmer credit delivery. Conceptually, group or cooperative organization of final credit delivery mechanisms represent a form of organization which embodies decentrali- zation of control and decision-making and incorporates local knowledge and responsibility. 3.13 The terms "cooperatives" and "farmer groups" cover a range of organi- zational forms. Farmer groups are usually informal in character, organized around a village, kinship relations, or common economic interest, and typically have about 10 to 20 members. These groups include farmers' associations, village societies, and peasant societies. If such groups grow to a much = 50 - larger size, they tend to lose their social cohesiveness. In some cases, the only real activity of the farmer groups is to be jointly responsible for the repayment of loans made to members, while the other functions involved with delivering credit are retained by the financial institution which provides the funds. More authority is transferred where the group is made resDonsible for the division of funds among members and the collection of loans. Still further develonment takes place if the group manages a ioint investment such as a tubewell. The group may also become a political organization for the ;ma11 farmprr wherebv he gains nower vis-a-viq larger farmers and narticripnst in the decision--making process of the credit institution from which the group nhotins itrs fiunds 3.14 Formnl nonperatives are usuta11v larger s-cale organizatinn.s hbuilt- around such functions as credit distribution, the supply of inputs, the manglopTnTnt oF inint investmients c,uch ns storage facilities or processing plants, or the marketing of output. Credit cooperatives are financial institutions w!mirhh usually hire their own professional staEff. Farmers ar requn-iraA to mnoe a financial comamitment to the institution through the purchase of equity shares and sometimes also through deposits= In some cases, the credit cooperative is restricted to credlit distribution and separate organizations are created to rnanage jinint investments or handle marketing; in other cases, one miltti-purpose cooperative is responsible for all of these functions. To be successful, cooprativesr must have-,. pa cetain mini mum size7 andA carry, otI~ a crtal- ,,,n4 , ,rni:au volume of operations. The trend has been toward an increase in the size of these ins ti4tutions for economic reasonc s a, u ly withi consequent adverse effects on their social cohesiveness. 3.15 Sometimes, informal farmer groups work in conjunction with formal fa.r.er cooperatives. Some responsiblite suc as Join signln for -loansI- --1 and collection of repayments are relegated to the farmer groups while others, ;)Uc-u as it:e cI& Lc Lt m--lerIs J an 7 1 Lie LIsUnLI oln ILoainI appLcatod LnU, ilady Ut: retained by the cooperative. In cther cases, farmer groups deal directly with L'iancial nstitutiLons of aI n0on-coouLt-puLerLLive charatL&LerL . ILL TLhailaald, for example, commercial banks lend directly to farmer groups, and in Western State oL Nigeria, farFmer groups Dorrow UirectCLy Lfrom tehe Agricultural Credit Corporation. 3.16 W4hen functioning properly, the use of groups and cooperatives as tne clhannel to deliver credit to small farmers has numerous advantages. Decentralization of the day to day aspects of management increases the credit programs' adaptability to local conditions and reduces the time required to process loan applications and make other decisions. Local knowledge can be used to assess both the risk of lending to a particular farmer and his invest- ment opportunities. This, plus group responsibility for repayment and equity participation in the cooperative, should reduce default. Furthermore, group- ing farmers raises the average size of loans, thereby reducing costs, and increases the political power of the small farmer. - 51 - 3.1 7 Th.e cooperatives considered b to he most success,full care those in Egypt, areas of India, Israel, Southern Brazil, Korea and Taiwan. Some have c.ombined info.mal v411-ae --rups with financial institutions operating one cooperative basis; this was quite successful done for a time at Comilla in nB anglad esht. Y.owever, in __. sa-t o U. alL1h6og - uccssful in deiern cre4.-dt A4 and other inputs to the farmer, the so-called cooperatives have operated more lik1e cer.trallze crdi burzuras .L.A.N L . A-LLL .U ze L cr AL.L. UUL=V&%_.C&%~...Les. J .10 ThILe systeLnL of cooperatives in L'Korea L a cU.Ls L L. AimLLUVt a.ll farmers are members, and almost all are small farmers. There is one insti- tution involved, the National Agricultural Cooperat've FederationL, Wllwhich is organized into three tiers, at the national, county, and village or township levels. Overall, this system can be judged to be quite efLicient, providing ancillary services as well as credit with relatively low administrative costs. Arrears have also been kept at manageable levels, averaging only seven percent of total portfolio in recent years. However, the system is tightly structured and highly centralized, and farmers have virtually no say in operations. 3.19 Institutions such as these are cooperatives only in the sense that a financial commitment in the form of shares or deposits is required of members. In other respects these institutions are quite simiiar to speciaiized agricultural banks and many suffer from the same problems. Cooperatives and farmer groups which are truly locally controlled have proven difficult to establish. Many societies are quite individualistic and small farmers every- where tend to be difficult to organize. Moreover, in many countries both the government and the wealthy farmers who are the local leaders have discouraged the growth of representative small farmer organizations. Cooperatives have tended to work best where there has been homogeneity of land holdings and tenure status among farmers, some grass-roots source of social cohesion, and a reasonable level of literacy. Grouping small farmers into cooperatives and farmers' associations has great promise, but the development of these organizations inevitably takes time, adaptation to local conditions, and requires a strong commitment from government. The Bank should foster the development of these conditions, but must avoid pursuing a single or rigid concept of cooperative organization since clearly many forms of association can be effective. Project Authorities 3.20 Special project authorities or project units established to execute various projects of regional integrated development, land settlement, crop development and irrigation have in many instances retained the responsibility of extending credit to participants for on-farm development and working capital. Often they had no alternative because there was no agricultural credit system or the existing one was unsatisfactory. In some cases, there were recognized advantages for credit to be closely associated with other services provided by the authority. These project authorities are Government departments, public boards or statutory corporations, funds set up by special - 52 - acts, or companies organized under special or general legislation. Project authorities have generally established a special unit staffed with the necessary expert staff to deal with agricultural credit operations. In most cases, but more especially in the case of authorities set up to develop the production of a specific cash crop or to carry out a regional integrated development project, these arrangements have achieved close integration of credit operations and ensured productive use of credit and satisfactory repayment records. Cooperatives and farmer groups are in some cases inter- mediaries between farmers and the project authority. There is a need though to develop apex institutions which can promote, oversee and service those enclaves waithin national programs. 3.21 The Bank Group experience with crop or regional project authorities is mostly in Africa because of the absence of suitable nationwide administra- tive mechanism for provision of various services and credit. Projects include the Lilongwe and Shire Valley Projects in Malawi, Tea Project in Kenya carried out by the Kenya Tea Development Authority, Wolamo Project in Ethiopia, Casamance Project in Senegal, Tea Projects in Malawi and Mauritius, and Cocoa and Palm Oil Projects in Ivory Coast. Non-Institutionalized Commercial Channels 3.22 In areas where institutional credit systems are quite inadequate, it may be possible to deliver credit through existing non-insitutional commercial channels. In most areas, these serve far more small farmers than credit institutions. By working through merchants, middlemen, and money lenders, existing credit agencies can increase the volume of credit available to farmers, and take advantage of the non-institutional lenders' greater flexibility, speed in lending, and lower administrative costs and default rates. Increasing the non-instituitional lenders' supply of funds may also tend to reduce interest rates. Lending through merchants also has possibili- ties since merch&nts, by splitting the overhead costs-associated with distri- bution over two functions, the delivery of goods and credit, are able to deliver credit in small amounts more cheaply than alternative channels. 3.23 There are also distinct disadvantages to such schemes. It would be difficult to prevent gross misuse of funds. Small farmers would usually require inputs and services other than credit, and it would be more difficult to coordinate tlhem when credit is distributed through non-institutional lenders. An attemnt to incoroorate non-instituitional credit into the Muda project in Malavsia has been abandoned. The effectiveness of public schemes to use non-inst-i-ittional sources is vet- to he demonstrited- Furthermore- the idea of using suchi conduits is politically unacceptable in many counrties beeause of their past exnloitRtion of farmPers On the other hand, eontrartors of farming operations such as custom tractor enterprises, suppliers of farm inniptcs and proDacsorQ or into-rm.Arinarios in the marketing nf farm nroduc,t-Q have provided an increasing amount of credit to commercial farmers. Under strict safeguards,l tie is ble. no realistic alternative is available. Institutional Commercial Channels 3.24 Commercial lending institutions represent another channel for lending to farmers. Many have an oxtensiup hranrh network snreading into rural areas. They are generally more efficient than Government agencies, and less prone to abuse and political pressre. TIn general, thyv are nrnnprlv organized, keep adequate records and are audited by independent auditors. They usually obtain better repay,Em..ent performan.ce by borrowers than do Government banks because of efficient mechanism and strict policies for loan collections. Because of the nature of the4- resources, commercial bar.ks aend, h-ever, to concentrate on short-term lending and to shun small farmers' credit because of costs. 3.25 In the Philippine, a private rural banking system developed from Government efforts and incentives to have local capitalists establish local banks aimed at meeting current credit requirements of rural communities, 4rcludin.g sml...l tar.m-.ers. Though the rural ,barnks aree partiall- y goover^nment- owned, they are nonetheless private institutions operated according to com__ _c _ L4O. fAt_4_A4t A4v4v. L L.± 4jC A. I1_C , UOr4 L.A 4SLLJU Shi cAA.LJ. .A C UCOA, 5t ^.' L.f .LS4 1 comiercIial princip a lesC| D. TraLs insituion hav e*ID,4-d LsteaLdVzOI.tCCz sUsS>D* - V S-aL z past 20 years, continue to expand their activities and now provide a significant portiLonLL of i'LnstiLtut'Lona.L creUiLt to agriculture L in the PIhilippines. Two Bank loans have already been channeled by the Central Bank through these institu- tLions fLor on-len[ding medLiu,I- andU Long-teru- to sL-uall far.im,ers, types ofL credit which were not extended previously. Rural banks lend to commercial farmers of Up to J50 hectaresC a sCj gndli"LcrllL poUt LoUn Ut Lle'Lr 'LeuLng hs 1nc LcUOlieu tIUJs w Lt farms of three to ten hectares. However, those with farrms of less than three nectares, wlho comprise 73 percent oL all 'armers adu cultivate 39 percent of the land, have received less than one percent of credit extended by the rural banks. To encourage these banks to lend to this class of Larmers, the authorities have had to introduce modifications which make the small farmer lending program of the rural banks similar to more traditionai pUDi1C credit programs for small farmers. Additional subsidies have been introduced for small farmer lending, including an 85 percent government guarantee of repayment on small farmer loans. The Philippine experience suggests that commercial type institutions of a specialized character, if they charge positive rates of interest and receive some government encouragement, can lend to medium-size farmers, but that in the absence of farmer groupings, heavy subsidies, or significantly higher rates of interest, they will not finance the smallest farmers. The rural bank experience is being replicated in Soutlh Vietnam. 3.26 In Thailand, private commercial bankcs have also lent to small farmers. This activity was concentrated principally in one institution, which loaned to farmer groups organized by the bank itself. It was only a break-even activity and ceased to expand several years ago. A state-owned general purpose bank, the Bank Rakyat, in Indonesia is providing credit to small farmers directly, and the recently nationalized commercial banks of India are also be,ginning to lend to small farmers. But as a general rule, to be successtul in lending to large numbers of small farmers, commercial institutions will need to rely on the same elements as public institutions, such as farmers' groups and government financial support. They also require appropriately trained mana2ement and personnel, bureaucratic decentralization, and some degree of insulation from political influence. 3.27 Because commercial banks constitute the core of financial markets in most developing countries, several governments, such as Bangladesh. India and Mexi,co, are now experimenting with the use of the commercial banking systems as a channel for reaching small farmers. Where commercial banks already have extensive networks of rural branches, the costs and delays associated with building new institutions are eliminated. But the costs to the banks to include small farmers rise appreciably. Commercial banks have been induced to lend to small farmers only through some form of government intervention, such as compulsory investment quotas or incentives such as special rediscount facilities and loan repnavment guarantees. The use of snpecial redisrount facilities or repayment guarantees for small farmer programs are preferable tn invpqtment uniotnaq in that thev are morp Pasqilv maecs qnerifir to smanll __~~ _ I___________ _---J S--- - - r -~-- ~ __ ~---~ farmers. 3.28 Bank Group lending for agricultural credit through commercial banks, via Central Bank or a sDecial fund, has proved satisfactory from a managerial and control standpoint and has been effective both technically and in reaching the cliontele nf lnrge hnrrntw.r.s rommerrial hanks hold littlo promise of dealing with small farmers directly, because of high administrative costs, lack nf hnrrnower clatgral and locational limitations which raot-rict acczss Government regulations in prescribing quotas for lending to agriculture and snpcial inepnt-iv,es fnr loans to sma ll farmers have not mnoiied thae romm.erninl banks substantially in this direction. Commercial banks are bound by the same kind of limitations as Agricultural and Development Banks and also would need to use intermediate channels such as cooperatives or farmer3' groups i-n order to reach sga4n4f4cnt n-m..ber of -.nl 1 fya-m 4. Guidelines for Delivery Sysems of Agriutrl Crei Larj_;r _FarrIer Chael 3.9 'it has been reco,.n,mended above para 2.4 andU 2.5) that credit programs for large farmers should be separated from those for small farmers. The basic principle be[hind lending to large farmlers should be essentially commercial and allocation of resources to large farmers should be made only if the opportunities they face are more productive tnan those wnicn can be founcl in other sectors of the economy. Large farmer programs should be expected to operate according to Dasic commercial principles, to cnarge realistic rates and achieve financial viability. -55 - 3.30 wnhere private Lanc'al iLnstLtutiLorLs are aduequate, they can ue useu to finance commercial agriculture. Unfortunately, in many developing countries, there is a bias on the part of commercial banks against agriculture, even extending to large-scale farming, and a preference to lend to urban based sectors such as trade and industry. To counter tnis bias, governments in some countries have attempted to channel the loans of the commercial banks into agriculture through compulsory lending requirements. The advantages of such an approach are deceptive; compulsory lending requirements reduce the ability of the banking system to allocate resources efriciently. wnere coupled with low interest charges for agricultural borrowers, this restults in low deposit rates, lessening the ability of the banking system to moDilize savings. Commercial banks should be encouraged to innovate and to move away from their traditional orientation of lending almost exclusively to trade and to large industry; but this is better done through persuasion, through preferential rediscount rates employed on a temporary basis to familiarize commercial banks with agricultural lending, or through the creation of new commercial institutions specializing in agricultural credit. Small Farmer Channels 3.31 Lending to small farmers is considerably more complicated than lending to large farmers: the large number of small-sized loans coupled with geographical dispersion make small farmer credit considerably more expensive to administer. A number of distributive systems have been tried including specialized agricultural banks, cooperatives, and commercial lenders, both organized and unorganized. 3.32 No one approach can be cited as superior to all others and to a large extent the choice of the best channel to deliver credit to small farmers will depend upon conditions specific to each country, or even to regions within a country. While different programs have been used in various countries, few governments have experimented with alternative delivery systems within their country. Because it is feasible to utilize alternative delivery systems at the local level while maintaining a single apex structure, experimentation should be encouraged to determine which systems are most compatible with local social and economic conditions, reasonable in cost and able to achieve the objective of delivering credit to those small farmers with productive investment opportunities. 3.33 Although much has yet to be done in experimenting with alternative channels f6r particular situations, certain principles have emerged from past experience which serve as guidelines for such work. These may be summarized as 10 basic points: (i) Accessibility. The branch office approach has not always proved effective in reaching a small farmer clientele for several reasons; small farmers have difficulty in reaching branches because of limited transportation; they are - 56 - unfamiliar with, and intimidated by, offices and office procedures; and branch staff tend to isolate themselves from farmers' problems. Consequently, it is necessary for the institution to go to the villages or even individual farms, to cooperative meetings or small informal groups to make the necessary linkages. The use of field agents who can visit small farmers in their own surroundings has proven successful. So has the use of mobile units or periodically opened village offices, as on market days. (iij Packaging. Credit alone is of little consequence. To be effective in increasing smallholder productivity it must come with details of proven new technology, and timely supply of inputs. (iii) Distribution in kind. This has proved to be the most effective means of delivering production credit to small farmers. Even for livestock it has p)roved to be worthwhile to do this.While this may seem excessively paternalistic, it is on the other hand completely unrealistic to expect people not used to having a cash surplus or to use of purchased inputs to automatically adjust to making the necessary transactions even if the required inputs were readily available. (iv) Timing. Especially when credit is provided in kind it is vital that inputs be made available when they are needed. Since smallholders can seldom provide adequate storage, too early delivery often results in losses and wastage of inputs such as seed and fertilizer. Too late a delivery in terms of seasonal requirements is, of course, pointless. Many smallho]der development projects have failed for want of the logistic support to get inputs to the farmer on time. (v) Selection. Smallholder borrowers should be chosen on the basis of creditworthiness, just as for any other borrower or credit pro-ram, but the appropriate criterion for assessing creditworthiness might differ. Three things seem important: the reputation of the individual within his community, the technical feasibility of thlie proposed investment in the context of his farm situation, and the expected cash flow that is generated by the investment. (vi) Individual liability and gro uj responsibility. While loan repayment slhould be the individual borrower's responsibility, and should be assessed in relation to his expected repayment ability, there is a strong sense of community in most rural - 57 - situations which can be used effectively to reinforce the honest performance of individuals. If transactions are made in public for all to see - quite the opposite of conventional financial practice - and the whole village or cooperative society made aware of the possible implications of an individual default (either in damage to their reputation or some form of penalty) group responsibility can become an important influence in ensuring repayment. (vii) Control. It is unreasonable to expect people with no knowledge of financial practices to suddenly conform to institutional lending regulations. Payments and repayments need to be scheduled to meet the liquidity shortage and surplus as it arises. If this is done, the supervision of even large groups can become an orderly activity. (viii) Flexibility. This is necessary to permit the rescheduling of loans in times of crop failure or other unexpected events. Often individuals' circumstances change, too, and they require some adjustment on the part of the program. There are many things that can go wrong in any program and these increase as the complexity or required precision of the scheme increases. Flexibility in lending criteria is also essential since no two communities or farmers are exactly alike. (ix) Continuitv. The failure of many programs to continue past a pilot stage of a single year or so, often leaving slow reDavments outstandine. encourages mistrust and poor discipline. Continuity of services is essential if programs are to have a nositive imnart. (X) An on-ended_ nnnrnach. Since progrnms need to he tailnrpd to meet local conditions, no fixed prescription can be set down. A sten-wise nroress of trial and error Jq neressary, beginning with a simple package and small numbers of hnrrowersq thpn rnrqRol-r1ntaina andi stthcQapi:nriv renlirnring this program to reach the whole target group. But an open-ended approach Is also needed because agricultural modernization is a continuing process. Just as in Western agricullture, we can. ex.pect each r. rour.d of technical innovation to require increased amounts of capital. It is unlikely that a single in4ectior. of A4+w411 n credit v-erca sufficient liquidity to make continuing modernization tolfprovtide f eius pcrogrnmi a shaould be planned so ali to provide for their continual adjustment and evolution. ANNEX TABLES ARI.TVTATTlNS FOR INSTITUTIONS MENTIONED IN ANNEXES Africa CADU (Ethiopia) Chilalo Agricultural Development Unit ADP (ana) Ag l Deve t Bank GMR (Kenya) Guaranteed Minimum Return (program) AFC (Kenya) Agricultural Fir.ance Grprat-3ion SOCAP (Morocco) Societe de Credit Agricole et de Prevoyance CNCA (Morocco) Caisse Nationale de rredit Agricole CLCA (Morocco) Caisse Locale de Credit Agricole l.vvRC fNlgera)4 V-f-_ 12fi4^n 1;P4nav%^, rf%v%%r!at-i%nn FAID (Nigeria) Fund for Agricultural and Industrial Development ABS (Sudan) Agricultural Barnk of Sudan BNT (Tunisia) Banque Nationale de Tunisie COOP (Uganda) Cooperative Credit System Asia ADBA (Afghanistan) Agricultural Development Bank of Afghanistan .AMUR (B.a1ngadesh) AgiuRu evlpen an.e eof Bamrngagdesh ADB (Bangladesh) Agricultural Development Bank BBfR (Bngladesh Rrishi Bar.k COOP (Bangladesh) Cooperative Credit Svstem IRDP (Bangladesh) Integrated RuralDevelopment * . KTCC (Bangladesh) Kotwali Thana Central Cooperative (association) PCSS (India) Primary Cooperative Credit Societies PLDB (India) Primary Land Development Bank BI.KAS (Indonesia) Acron ym for "Birubi.gan Mssal" me,in- "H-a- Guidar,c" ACBI (Iran) Agricultural Cooperative Bank of Iran ADFI (Iran) Agricultural Developuent Fund of Iran ACC (Jordan) Agricultural Credit Corporation BCAIP (Lebanon) Lrebanese Creditcl Bank, for Agricultural and Industrial Developmen a_.... v..Ve...,.s Uv- ;o "L m - Insl J.U ^L LU L Ld AU L I^U -LL J. HeUU--t:L0 ML1L BAAC (Thailand) Bank for Agriculture and Agricultural Cooperatives SCGP (Turkey) Supervised Creditr DPro ramr NACF (Korea) National Agricultural Cooperative Federation TRAR (Turkev) Tuvirkiah Depublcan Agricultural Ban'k 1 flL 5. 4. t_ __ ACAR (Brail7 ) Associacao deCreditoe Asistencia Rural INDAP (Chile) Instituto de Desarrollo Agropecuario INCORA (Cn]ormbia) InstitutoiColomblanro de la RCeform Agraria BCR (Costa Rica) Banco de Costa Rica BNCR (Costa R1ra) RBnan Nacionni Ao ralta Rica DACP (Ecuador) Directed Agricultural Credit Program DAPC (Et-ctuainr) Diret ed Agricultural Production Credit (pLugL-) ABC (El Salvador) Administration de Bienestar Campesino BNF (Honduiras) Rnn^^ National de Fom#ento FONDO (Mexico) Fondo de Garantia y Fomento Para la Agricultura, Ganaderia y Avicuitura Guarantee and Development Fund for Agriculture, Livestock and Poultry NBN Rur. Cred. (Nicaragua) National Bank of Nicaragua Rurai Credit (program) NBN (Nicaragua) National Bank of Nicaragua ADBl (Peru) A^rIcultural Development Bank Mnnex TLable i: W?Ln1Ui1ON?A¶j L1?oWMU FOR AunILkunE (US$ million) Dollars Per Capita Year of Loans New of Rural Observa- Country w Outstanming L'oans ropulavton tion Africa Ethiopia 18 6 1 1970 Ghana 19 6 b 1971 Kenya 131 12 197( Mlorocco 130 65 13 1971 'Tunisia. 53 15 20 1970-71 Uganda (Co-op only) 3 1971 Asia =g anistan 2 1 1 1971 Bangladesh 130 4.3 ] 1972-73 India 2,400 1,380 3* 1971 Indonesia (BRI only) 72 1* 1971 Iran 159 127 9 1970 Jordan (ACC only) 17 2 17 1971 Korea (NACF only) 236 206 55 1971 I'ialaysia 200 .20* 1.971 Pakistan (>,.est) 33 26 1967-68 Philippines 523 443 b 1971 Sri Lanka 30 9 74 1970 Taiwan 409 225 3 1971 Thailand 73 L2 20 1970 Turkey 414 2* 1967 Vietnem 32 1971 Latin America Argentina 555 111 1968 Bolivia 15 5 8 1971 Brazil 1,500 ho* 1969 Chile 261. 106* 1969 Colombia 416 319 52 1970 Costa Rica 110 126 1968 Dominican Republic 57 24 1968 Ecuador h8 13* 1968 El Salvador 78 36* 1970 Guatemala 52 18 1968 Honduras 59 35* 1971 Mexico 1;671 84, 1971 Nicaragua 1()0 77 1970-71 Panama 23 I1 1067 Paraguay 33 22 1968 Peru 16n 25 1967 Uruguay 18 36 1967 Venezuela !4.R8 179 1968 *Per capita vale of' new loans extended dnrina the year rather than the per capita debt outstanding. (cont1d) Annex Table 1 Paee 2 NOTE: Loans outstanding represent the end of the fiscal-year.portfolio of the-lending institutions.- As-few credit agencies-write off.bad,debts. this f-iigure is inflated-by2) 72) (Nos.) (') East Africa 0-5 ha 17.7 30.5 483,580 97.9 oi- which Ethiopia 1/ (5.2) (9.0) (400,000) (81.0) 5 1 ha=10 ha - - - - 10.1 ha-100 ha 4.2 7.2 9,150 1.9 of which Ethiopia (2.3) (4.0) (350) (0.1) Over 100 ha: Private 11.4 19.7 855 0.2 of which Ethiopia (3.8) (6.6) (245) 2/ G-overrntent 21.8 37.6 121 7/ Cooperatives 2.9 5.0 43 2 Subtotal 58.0 100.0 493,749 100.0 West Africa U-5 la 8 l. .-1AA 5 24 ., 5.1 ha-10 ha 13.7 43.5 70,000 17.7 10.1 ha-100 ha 9.4 29.8 225,000 56.9 Ovpr 100 ha: Private - - - - Government - - _ _ Cooperatives - - _ _ Subtotal 31.5 100.0 395,500 100.0 Asia 0-5 ha 206.4 57.6 320,090 79.5 of which india (184.6) (51.5) (295,800) (73.5) 5.1 ha-10 ha 40.5 11.3 38,610 9.6 of which Inda (32.8) (9.2) (36,000) (89) 10.1 ha-100 ha 106.8 29.8 43,870 10.9 of which India (58.4) (16.3) (17,600) (4.4) Over 100 ha: Private 1.7 G.5 63 2! Government 3.0 0.8 9 2/ Cooperatives - - -- Subtotal 358.4 100.0 402,642 , 100.0 EMENA 0-5 ha 9.1 4.8 12,000 12.3 5.1 ha-10 ha - - - - 10.1 ha-100 ha 104.0 55.0 79,815 81.6 Over 100 ha: Private 49.4 26.1 5,696 5.8 Government - - - _ Cooperatives 26.7 14.1 254 0.3 Subtotal 189.2 100.0 97765 iO0.0 LAC 0-5 ha - - _ _ 5.1-10 ha 1 _ _ 10 1-100 ha 1 7.3 44,025 60.2 of which Mexico (117.0) (31.9) (32,900) (45.0) Over 100 ha: Private 218.8 59.7 29;075 39.8 of which Mexico (101.7) (27.8) (11,100) (15.2) Government - _ Cooperatives - - _ Subtotal 366.5 100.0 73,100 100.0 All Regions 0-5 ha 241.6 24.1 916,170 62.6 5.1-10 ha 54.2 5.4 108,610 7.4 10.1-100 ha 372.1 37.1 401,860 27.5 Over 100 ha: Private 281.3 28.0 35,689 2.5 Government 24.8 2.5 130 2/ Cooperatives 29.6 2.9 297 2/ Total 1,003.6 100.0 1,462,756 100.0 *Based on anticipated results as noted in Appraisal Reports. 1/ Ethiopia Agr. Minimum PackaRe Proiect. 2/ Less than 0.1%. Annex Table 7: BANK GROUP AGRICULTURAL CREDIT OPERATIONS, BY Cr J.PNY AND PER C.APITTA NP TRIM rv 19i48-10973 (US$ million) Country and GNP Group FY 19-196 FY 1964-1968 FY 1969-1973 DC ;A i: u L 7u 2%4 Lv . Les ULan .L1, W Botswana 1.7 (1) E 4-opi a 37 .2 (t) Kenya 10.2 (2) 3.6 (1) 9.6 (2) rL1is41Iasy R(i Malawi 9.7 (2) 17.1 (2) Sui--, t(1) 16.3 (2 Tanzania 5.0 (1) 39.6 (4) VT .a 3.4 (1.I) 7. (2)o U gtLU.L ..L %I'.a\ Zaire_ _ _ _ _ ___ _ _ _ _.5 ( . 4 Sub>Do 'P.) - 0 (2 7 ( 19. (1 Rhodesia 5.6 (1) $376-$700: .L,aIn Dla _ _ _ _ _ _ _ _ _ _ _ _ _ A t . % 1 11 7 _ _ _ _ _ _ _ _ . a I. n I Total East Africa 15.8 (3) 21.7 (5) 153.8 (21) West Africa Less tnan $150: Dahomey 6.1 (1) Gambia 1.3 (1) Niger 0.6 (1) Nigeria 7.2 (1) Upper Volta 6.2 (1) Suvbtotal 2T.7 (5) $151-$3 75: Ghana 9.8 (2) Ivory Coast 7.5 (1) Senegal 17.9 (3) Siera Leone 4. l') Subtotal 39.5 Total West Africa 60.9 (12) NOTE: Figures in ( j are the number of operations. (cont'd) Annex Table ? Page 2 Country and CNP Group FY 1948-1963 FY 1964-1968 FY 1969-1973 Amount AM Amount T() Asia Less than $150: India 42.2 (3) 306.9 (1) Indonesia 19.9 (4) Pakistan 30.0 (1) Subtotal 2.2 (16 $151-$375: Korea 17.5 (2) PapuaANew Guinea 11,5 (3) Philippines 5.° (1) 31, 6l Subtotal 370 (1) 60.6 $376-D700: China 13.7 (2) Malaysia 25.0 (1) Subtotal 13.7 (2) 25.0 (15 Total Asia 60,9 (6) 442.4 (25) BMENA Less than $150: AfghanistAn l. 0 (2) Yemen P.D.R. 3.5 Jordian 6.0 (2) Morocco loO (I) -34. (1)~ Tunisia 19.2 (3) Subtotal 20.0 (1) 16.0 (3) 9B,7 (7) $376-$?00: Yugoslavia 31.0 (1) Finland 20.0 (1) Icu±lnd 2.L4 k2) Ireland 25.0 (1) israe'L Ju. ) Spaine 20.0 (1) Subtotal 2.4 (2) 25.u 0 v Total RENA 22.h (3) 16.0 (3) 257.7 (17) (contfd) Annex Table 7 Page 3 Country and GNP Grou FY 1948-1963 FY 1964-1968 FY 1969-1973 Amount ($) Amount-M$ Amount ($) LAC Less than $150: _ $1514$7 Bolivia 2.C (1) 8.2 (2) Colombia 10.0 (2) 16.7 (1) 43.4 (3) Dominican Republic 5.0 (1) Ecuador -4.0 (1) 16.8 (3) Guatemala Li.0 (1) Guyana 1.3 (1) 2.2 (1) ilonduras 2.6 (1) Paraguay _.0 (1) 11.1 (2) 8.6 _ Subtotal 16.3 (W 33.8 (5) 90,v (Z5 S376-VO7O Brazil 40.0 (1) 116.7 (4) Costa Rica 6.5 (2) 3.0 (1) 9.0 (1) Jamaica 3.7 (1) Mexico 25.0 (1) 250.0 (3) Nicaragua 2.7 (2) PA-M 1<.o (3) 26.0 (2) Subtotal 24.2 t759. 379.4 (9) Over $700 ,^^gentina 15 3~l~ (1) Chile 2.5 (1) 19.0 (1) Panram.a 1.2 (1) 147 (1) Trinida.d Tobago 5.0 (1) Uruuay 7,0 (I 12.7 (1) 21.5 (3) Venezuela _ 11.0_(1) Subtotal 10.7 ( -4 52.n- (4) 3?= (5) TV-o+. T At' 51. 2 (A 17Q.R (I)) 5c07, (97) All Regions ( A 4 (%) Less than 10.2 11.4 o3.y 22.9 5>5.5 37.7 $1514375 41.9 46.9 54.8 19.7 289.6 20.4 W376-$700 214.2 27.1 107.7 38.7 469.9 33.0 Over $700 13.1 14.6 52.0 18.7 127.2 8.9 Total All Regions 89.4(20) 100.0 278.4(28) 100.0 1,1422.2(102)100.0 NOTE: Figures in ( ) are the number of operations APnan Table 8 A-oex Tabla A: CONTRIBUTION TO PROJECT COSTS BY ThkNK GROUP AGRICULTURAL CEREIT OPERATIONS, BY PER CAPITA CNP CrNTRY CATEGORIES, FY 1948-1973 (USS olilion) _ _ ~ ~ ~ ~ ~ ~ ~ ~ ~ - n 1S48___3 _ __ n 1'6-66 _____________- FT 1597 - ProJ3ct _ LaCratdrt Loall Coet t____ Project _ Total I Loee C6t Prolect 70=i _0 Credi local Cost Par Capit CtP A' 4 Ot RoJeet 4 S or t Sa of Project of Borroing C ootriea Total Local Cost Amoint Total Project Amoont Local Coat Total Local. Cost Amoont Total Project Amount Local Coat Total Local Cost Amount Total Project Amount Local Cost - ~ ~~~~~~~~ ' ($, )- _ (f) ( MS) ($) (j () 1) ,iL (s) (S (s) (%) _ ($1 ($) ($) (%) - ($) - () ,__ Less than $150 24.7 14.5 10.2 l1.3 *.- _-_ - 153.7 103:.8 63.9 41.6 19.0 18.3 91:2.4 587. 535.5 58.7 227.5 38.7 $151 - $375 102.0 73.9 41.9 11.:L 13.8 113.7 102.1 63.7 54.8 53.7 16.9 26.6 551.0 .318.1 289.6 52.5 62.5 19.7 $376.- $700 49.9 25.7 2h.2 48.5 -_ * _ 228.6 1&65.1 107.7 47.1 44.8 27.0 1,154.3 861.5 469.9 40.7 174.tl 20.3 mer S700 27.8 1h.7 13.1 47.1 -.- -.- 1h3.5 97.0 52.0 36.2 5.5 5.7 3fl.l :.j 'i27.2 33.4 21.5 7.8 Total 204.4 1288 89.4 43.7 2.3.8 11.7 627.9 4-31).6 278.4 14.3 86.2 20.0 2,999.8 2,042.5 1,422.2 47.4 486.31 22.9 ' World Benk Atlas 1972. The coatrias wtt,h edit plroj aet. guar stach be cataorpy are shown tin A-ax Table 7. 1N f-i CD Annex Table 9: INTEREST RAIES TO FARMERS BY SOURCE OF LOANS (percent per year) Nominal Rates Real Rates Commercial Commercial Country Institutions Lenders Institutions Lenders Africa Ethiopia 12 70 8 66 Ghana 6 70 0 6h Ivory Coast 10 150 6 144 Kenya 7 7 Morocco 7 5 Nigeria 6 200 -2 192 Sudan 7 120 7 120 Tunisia 7 1 Uganda 12 1 Asia Aganistan 9 33 Tndia 10 2 0 1 Indonesia 14 40 3 28 Tran 7 5 Jordan 7 20 2 15 Korea 16 60 5 4 IIalaysia 18 60 16 58 Pnk; ct-qn 7 in 97 Philippines 12 30 6 22 Sri LaTna 12 50 4L, Taiwan 10 3 Thail-nd 11 30 9 28 Vietnam 30 48 2 20 Lati-n Aymneri ^a Bolivia 9 100 5 96 Brazil 15 60 -7 39 Chile 111 82 -16 52 C'ol ,Lvia 12 41 3 Costa Rica 8 21 4 20 El Salvador 10 25 8 38 T~~~~~~~~~~~~~~~~~ rouLuras 7 i4° 6 5 Mexico 10 60 7 Nicaragua 1.3 8 Peru 12 5 NOTE: The nominal interest rate reported for institutions represents an average rate for different types of loans and various institutions. One uniformity is that private institutions tend to charge about 3 percentage points higher than public institutions. The informati.on on the rates cnarged by commercial lenders (cont'd) Page 2 conl UIi±y Uo n<- tul1 cb ta'rLVV. ais uUaiv At besLths ji±ALLUMU.LA.JL w-s based on smple surveys; often it represents only hearsay evidence authors chose to report. In Pract,ice miolS t commercalaCL jloans a sliUL , onIe tU JLL.-UW IJU1L be.gL1L typical. Multiplying monthly rates by twelve exaggerates both the cost to the borr-ower anu the lincot ,o t, iIe luenduerr. ConversLon from nominal to rreal rates was done by deducting -for, the nominal rates, the average rate of inflation between 1967 and 1970, except in a few cases such as Indonesia where the inflation rate betwween 1967 and 1970 clearly did not reflect current levels of inflation. A.-unex . able 10: B A NK CROUP AGRiPITMITMLTh rIPFfjT'IT OPFRPTTfl?_ BY MAJOR END USE AND BY REGION, FY 1948-1973 (US$,t I 1 -n) East Africa Wdest Africa Asia EMENA LAC TOTAL No,o No No. f No=of Nooof No.of Major End Use O Amt. Op. mt. Op, Amt mt. P Amt. Op. Amt. W) T (Nosi (N Tos7 '.No ( (Nos.) GenerEl Ac, Credit FY 1914_-63 1 5.6 - - _- - 1 1.3 2 6.9 FY 1964-68 2 8.6 - - _ _ 2 13.0 2 18.0 6 39.6 FY 1969-73 2 11.5 3 114.8 107.1 6 76.7 3 29.7 19 2l0.1 Sub-total 5 -T. 7 - ,T T17.T 7' 4190 2 d97 6 9 2 26.6 Livest.ock FY 1948-63 - - - -- _ _ 2 2.4 1 7.0 3 9.4 FY 1964-68 - - _ __ _ _ - - 10 145.8 10 145.8 FY 1968-73 8 41.9 - -- 3 16.1 4 75.0 19 373.6 34 506.6 Sub-total W Ll-9 - __ 3 IbI -6 77Th 3C- 526Th 47 661.o Crcp Development FY ]9L0-63 - - - -- - - - - - - - - FY 1964-6E 1 3.l -7 - _ - _ 1 3.4 FY 1969-73 4 35.3 5 35.5 3 14.3 1. 20.0 - 13 105.1 Sub-totAl 5 T 35,7 -5 3 .3 1 20.0 - - 17 140.5 Irriga .ion FY 19118-63 - - - - - 1 20.0 - 1 20.C FY 19614-68 _ - - - 5.2 I 3.0 - - 2 0.2 FY 1969-73 - - - 14 lL1.0 - - 4 141.0 SDubu-totai. I - - -_5 t-y--c4 --- 2 - Farm Machinery -) I .-.1 5,6 - _ - _ _ _ 12 LL.9 13 7 4U FY 1961-68 - - - - 3 42.0 - - - - 3 12.0 ,Fv l969-73 2 16.3 = - - - - - , 0 Sub-total. T 7T7 - - - T34 - - 12 32.9 aT I5737 Fi-sheries FY 1940-63 - - - - - FY 1961-6 - - - - 2 13.7 - - - - 2 13.7 FY 1969-73 - - 1 1.3 2 15.1 2 5.5 1 5.3 6 27.2 Sub-total - - 1 1.3 4F 2757 2 -77 1 -.3 --7- 60.9 Agro-Business FY 1948-63 .Y 19614-68 2 1.0.7 1 3.7 3 26.0 3 60.5 2 61.0 11 1E4.9 FY 1969-73 '--7r 10.7 1 I.7 3 Z 3 60.5 2 pi. 1 II.9 Sub-total I j.I -'. j 2 1 L.L .LJL4 7 Integrated Ag. Devt. I 0i; A _ _ _ _ _ _ _ 1 * 6 FY 19168-63 2 9.7 - _ _ _ _ _ 2 16.0 4 25 7 WY 196LL-68 3 38.1 2 5.6 2 30.0 - - 2 114.8 9 RR.< FY 1.969-73 -- - * -2 30.0 - - Sub-total Forestry FY 19148-63 FY ±90/L-OO - - - 2Q.0 _ _ 1 20.0 FY 1969-73 _ - _ - - 1 200. Sub-total m 29 19].3 12 6n.9 31 5C3.3 3 7297.1 5 731 15 -0 1790.0 Percent of Tota. 19 8 21 15 37 1in (No. of Op.) Percent of Total 11 3 28 17 ltl 1OC (Amount) ,Lnaex Table 11: DURATION OF LOANS MADE BY SELECTED INSTITUTIONS (percentages) Duration Less than From More than outr.t- Institutions 2 years 2 to 5 years 5 years Africa E¶T-1op -a All 100 Kenra All 26 31 L3 Morocco All 73 27 i i.q=n BNT 62 38 Ugazda COOP 100 Bangladesh ADB 48 8 TIna COOP 76 24t IrTr. ACBI 30 70 ADFI - in go Jordan. ACC 25 75 Korea. NACF 90 10 11alaysia BPM 60 0 Pai'-stan. COP 88 12 Philippines Rural Banks 93 7 (Z-; T .t.- All 9 ; s. Taiwan All inst. 50 28 22 .W^alnd a-; All 4 .s.+ 71 Jd.L J..Ci-aL 'A4 JM.L.J A.4. '* - A Turkey SCP 23 53 2b V J.'Z UIACV-W R r1U. a."L JJ.AJ Latin America Dr,azi- r fzeU. JD C 78 1°S Bolivia Agr. Banks 95 5 Chile Ii'luAr P.J Colombia Gaja Agraria 42 39 19 Costa Rica State Banks bO 60 Ecuador DACP 100 El Salvador All inst. 964 Honduras BNF h3 37 20 Mexico FONDO 90 5 5 Nicaragua NBN 82 18 NOTE: This table shows for specific institutions the cumulative distribution of loans by duration. In most short term production credits constituted the bulk of all loans. In some cases the institution initially made longer term loans but uinpaid loans were usually rescheduled for one year. Thus, in part the short term nature of the portfolio reflects the degree of rescheduling as well as the emphasis on production credits. Annex Tahle 19: MAqL1P S OF T.AAT TTNX OT.TWOTrW.?V fP 7.ST.VT'!n T(V.Tr1TCTTIT (percentage) A ars 4to rrA- rs Country Institations Portfolio Rate Africa rsu:~~~~~~~~.,~r _:. -, **4CAZ1J 50 Ivory Coast BNDA 15 Kernya GM4 25 33 t**AFC 51 36 Malawri *Wilongwe - 2 Niger *CNCA 11 29 Nigeria WSACC 52 80 FAID 95 Morocco SOCAP 5° **CNCA 13 5 Sudan 0OP 26 ABS 13 Tanzania iHNDCA 28 50 Tunisia **BNT 66 50 Local Credit Unions 50 Uganda Co-op Credit Scheme 10 Asia Aignanistan **ADBA 37 77 Bangladesh AB 76 IRDP LO India PCCS 34 7 **PLDB 12 20 Iran ACBI bt Jordan *',ACC 41 82 Korea ,-*MAFC 7 15 Malaysia BPR 6 21 Pakistan 4*ADB 36 65 Philippines **RRural Banks 20 18 Sri Lanka New Credit Scheme 50 h1 Thailand BAAC .5 Turkey ABT 29 L3 Vietnam Rural Banks 5 Latin America Bolivia **Agr. Bank 1 68 Chile INDAP 16 60 Colombia 'H*Caj. Agr. 19 *i.INCORA 4 16 Costa Rica BNCR. BCR 35 El Salvador ABC 37 81 Honduras BNF. Sup. Credit 10 18 Jamaica ADB 31 10 Peru Plan Costa 33 '*-BFA 30 * The arrears rate is equal to 100 minus the repayment rate. (cont'd.) ..' Institutions involved in Bank Group projects. Annex Table 12 Page 2 NOTE: These measures have various shortcomings. Most agencies consider rescheduled loans as having been repaid. A low ratio df arrears to port- folio may-not mean much when loans are expanding rapidly and not yet due while at the same time the repayment rate on previous loans is poor. Annex Table 13: AIEMNISTRATIVE COSTS FOR SELECTED INSTITUTIONS Cost as Cost as a Percent a Percent of of Total Countr Institutions New Loans Resources Africa Lhana ADB 10 10 Ivory Coast CNCA 9 Kenya *AFC 3 Morocco *CNCA 10 3 Senegal BND 3 Uganda Cooperatives 50 Asia Bangladesh KTCC 17 10 BKB 3 India *LDB 3 Indonesia BINAS (improved) 25 Jot-dan *ACC 30 3 Korea *NACF 6 h Lebanon BCAIF 3 Malaysia BPM 20 Pakistan *ADB 3 Philippines *Rural Banks 5 5 Thailand BAAC 13 8 Turkey SCR 5 BAT 6 Taiwan Farmers Association 2.5 Coop. Bank (2.5) Land Bank (1.5) Latin America Brazil AGCA 10 Colombia -,*INCORA 10 7 Costa Rice BNCR 7 3 Ecuador DAPC 4 El Salvador ABC 16 11 Mexico *F0NDO 3 1 Peru ADB 6 * Institutions involved in Bank Graop projects. NOTE: Capital and to the extent possible supervisory costs have been excluded from the cost information in this table. However. it was not possible to get comparable figures for different institutions. The very low cost figures re- ported by such institutions as the KTCC in Bangladesh reflect only the cost of the final lender and not that of the entire agricultural credit system. On the other hpnd- institutions with very hiah tnqt. fi;rrea nre n-rohnh1v nronAi8- ing farmers more services, the cost of which it was not possible to eliminate with thpe Yvail be data.- Ot.hern reasons for high 'csts insom inti ii+Ai ns are that the programs are new and of small size but have alreadty hired the staff that will en',be! them. to eNr an T.e B1RAK -in Malaysia is such <.n insti- tution. Annex Table 14: BANI GROUP AGRICULTURAL CREDIT OPERATIONS, BY LENDING CHANNEL TO ULTIAkTE BORROWER, FY 1948-19/3 1FY 1948-63 FY 1964-68 _ FY 1969-73 No. of Operatione Amount No. of Operations Amount No. of Operations Amount (Nos.) (MA ) (¶c) (Nos.) (%) ($) (%) (Nos.) ('A) ($) ('A) Coaercial Channel 4 20.0 14.7 16.5 6 21.4 101.0 36.3 27 26.5. 546.5 38.4 Agriculitural Banks 9 45.0 53.0 59.3 15 53.15 129.4 46.5 33 32.3 354.9 25.0 Cooperatives - - - - - - - - 1L 10.8 266.9 18.8 Developmnent lianks 3 15.0 12.1 13.5 5 17.9 39.1 14.0 17 16.7 117.4 8.2 Project Authority, Ministry 4 20.0 9.6 10.7 2 7.1 8.9 3.2 14 13.7 136.5 9.6 or Special Entity 1/ ______ Total 20 100.0 89.4 100.0 28 100.0 278.4 100.0 102 100.0 1,422.2 100.0 Average Size of Operation 4.5 9.9 13.9 1/ British Guiana: Credit Corporation; Korea: Dairy Beef Company; Malagasy: Ranclh Stalte Farm; Spain: Instituto de Credito a MEeV.c YAed Sudan: lMechanized Farming Corporation; Zairea: The National Ranching Development Authority. 115