managing credit risk in rural financial institutions v2

8/7/2019 Managing Credit Risk in Rural Financial Institutions v2

1/36


2/36

Mark Wenner is Senior Financial Specialist in the Rural Development Unit. Sergio Navajas is Microenterprise Spe-cialist in the Micro, Small and Medium Enterprise Division. Carolina Trivelli and lvaro Tarazona are researchersat the Institute for Peruvian Studies (IEP) in Lima, Peru. The authors are grateful for the valuable comments re-ceived from Juan Buchenau (consultant), Calvin Miller (FAO), and Dieter Wittkowski (SDS/MSM). The authors

express their sincere gratitude to all the institutions and individuals who participated in this study. Special thanks,however, are extended to Fernando Pea, Sergei Walter and Luis Morales from Banrural S.A.; Sammy Calle andLuis Lamela from CMAC Sullana; Elizabeth Ventura and Juan Meza from EDPYME Confianza; and Hugo Vi l-lavicencio and Guillermo Caal from Fundea. In addition, the authors are very appreciative of the efforts of JorgeLuis Godnez who provided invaluable help in designing and implementing the web-based survey utilized for thisresearch, Raquel Trigo who helped us to identify potential financial institutions with rural portfolios and to managemailing lists, Raphael Saldaa (IEP) for his invaluable programming skills, and Margarita Reyes who provided ableproduction assistance. Lastly, the authors would like to acknowledge the financial support received from the Norwe-gian Trust Fund admin istered by the Bank.

The opinions expressed herein are those of the authors and do not necessarily represent the official position of theInter-American Development Bank. Permission is granted to reproduce this paper in whole or in part for noncom-mercial purposes only and with proper attribution to the authors, the Sustainable Development Department and theInter-American Development Bank.

Publication of the Inter-American Development Bank, May 2007.

Manager, Sustainable Development Department: Antonio VivesDeputy Manager, Social Development and Governance Subdepartment: Marco Ferroni

Chief, Rural Development Unit: Csar FalconiChief, Micro, Small and Medium Enterprise Division: Alvaro Ramrez

Additional copies of this report (Ref. No. MSM-139) can be obtained from:

Micro, Small and Medium Enterprise DivisionSustainable Development Department

Inter-American Development Bank1300 New York Avenue, N.W.Washington, D.C. 20577


3/36

Foreword

Adequately managing credit risk in financial institutions is critical for their survival and growth. In thecase of rural lending in general and agricultural lending in particular, the issue of credit risk is of even ofgreater concern because of the higher levels of perceived risks resulting from some of the characteristicsof rural dwellers and the conditions that they find themselves in. More extremely poor people tend to livein rural than in urban areas. In addition, fewer people are able to access basic infrastructure services andthese tend to be of lesser quality or to be less reliable than in urban areas. Rural residents tend to be lesseducated, more often than not they have insecure land tenure, and they live farther apart than urban popu-lations. Most importantly, agriculture, the mainstay of most rural economies, tends to be subject to price

volatility, weather shocks, and trade restrictions. As a result, financial institutions that are active in ruralareas are likely to face an elevated level of credit risk and need to manage it well. The lack of good riskmitigation techniques and high transaction costs can discourage formal financial institutions from enteringand serving rural areas.

The purpose of this report is to review common credit risk management techniques used in a sample ofLatin American financial institutions with agricultural portfolios, identify the factors that contribute tosuccessful credit risk management as measured by several key financial performance indicators in orderto assist donors, governments, and owners of financial institutions to promote and adopt the most efficientand robust techniques. The ultimate aim is to make financial markets more inclusive and sounder.

The report also examines the results of a survey of forty-two rural financial institutions in Latin Americaand provides a detailed analysis of four intermediaries, two in Peru and two in Guatemala. It ends with aset of recommendations on how to improve credit risk management capabilities.

We are confident that owners and managers of financial institutions with agricultural portfolios, donoragencies, national governments, and other stakeholders will find this report informative and helpful for

making decisions and strengthening institutions.


4/36

Contents

Executive Summary i

Introduction 1

Review of Typical Risks Faced by Financial Institutions and

Techniques to Address Them 3

Summary and Description of Survey Results 9

Summaries of Case Studies in Guatemala and Peru 17

Conclusions and Recommendations 24

References 27Annex A: List of Participating Institutions in Survey 29


5/36

Executive Summary

Rural areas lack banking services due, in largepart, to perceptions of high risks and high costsof delivering financial services. In Latin Amer-ica, it is estimated that less than five percent ofrural households have access to formal credit.Even though agriculture is declining in eco-nomic importance and nonfarm activities arebecoming more important over time, agriculture

remains the main livelihood activity for many.Agriculture, however, is inherently more riskythan other sectors due to its vulnerability to cli-matic shocks, commodity price volatility, andtrade restrictions. In the current context of ongo-ing globalization and the quest to reduce ruralpoverty, agriculture will have to maintain andimprove its competitiveness. Ready access toagricultural finance is one of the main ways of

improving agricultural competitiveness. There-fore, it follows that lending technologies and, inparticular, rural credit management techniquesmust improve.

This report examines a sample of forty-two fi-nancial institutions in Latin America that haveagricultural portfolios, and identifies their prin-

cipal perceived risks, how they assess and man-age credit risk, and how effective they are in theprocess as measured by key financial perform-ance indicators (such as asset quality, portfolio

th d fit i )

Excessive provisioning is being used toabsorb and internalize risks.

The largest challenge for expanding credit inrural areas is that few institutions are transfer-ring credit risk to third parties. In developedcountries, massive expansions of credit havebeen due in large part to the introduction and

wide diffusion of risk transfer techniques (suchas insurance, securitization, derivatives, swaps,etc.) and the wider acceptance of different typesof collateral (inventories, accounts receivables,warehouse receipts, etc.). In the sample sur-veyed, the most common risk transfer instrumentavailable and used (albeit only by 25 percent ofthe respondents) is publicly financed guaranteefunds, which have historically been plagued with

problems such as high costs, limited additional-ity, and moral hazard. In order to introduce someof the other risk transfer instruments more com-monly found in developed financial markets,investments will be needed to reform andstrengthen the insurance industry, capital mar-kets, credit bureaus, commercial codes, securedtransaction frameworks, and information disclo-

sure rules.The implications of using these credit risk man-agement techniques are many. First, credit

l ti t h l i i d


6/36

present, a common set of credit evaluation prin-ciples seem to be widely applied:

Employ well-prepared staff that has somebackground in agronomy.

Use staff performance incentives to promotea sense of responsibility and to reward re-sults.

Gather and use copious amounts of informa-tion on character, managerial ability, reputa-

tion for repayment, and financial viability toidentify good borrowers.

Rely principally on cash flow and sensitivityanalysis to determine repayment capacity.

Give preference to households with diversi-fied streams of income and that are some-

what insulated from weather risks (largerhomesteads, fragmented plots in differentmicroclimates, and those that use of irriga-tion).

Use repayment incentives to avoid strategicdefaults.

Monitor clients closely.

In conclusion, most institutions surveyed sawmarket opportunities in rural areas, and the mostsuccessful institutions were expanding their ag-ricultural portfolios and generating profits. How-ever, much can still be done to improve creditrisk management by improving the feasibility oftransferring risk to third parties.

The report makes several recommendations fordonors and governments. The preferred or bestoption is to provide support to rural institutionsthat meet minimum scale requirements that

would permit easy diversification of credit risk,and help them to expand and innovate. In coun-tries where these types of rural financial institu-tions are absent, the second best option would beto assist those institutions that have a clear stra-tegic commitment to the ruralsector as well ascompetent management to upgrade their tech-nologies, diversify, and introduce risk transferinstruments. The third best option would be topromote mergers and acquisitions amongsmaller institutions so they can reach a largerscale. The fourth best option would be to pro-mote value chain financing, since many of thecredit risks are attenuated by participation in achain.


7/36

Introduction

Taking credit risk is part and parcel of financialintermediation. Yet, the effective managementof credit risk by financial intermediaries is criti-cal to institutional viability and sustainedgrowth. Failure to control risks, especially creditrisk, can lead to insolvency. However, too often,the mere perception of high credit risk can dis-suade financia l intermediaries from entering aparticular market segment when a large contrib-uting factor to that perception may be lack ofadequate credit risk evaluation and managementtechniques. This seems to be the case with ruralfinance, especially lending to sma ll- and me-dium-scale agricultural producers. If financialinstitutions do enter rural areas, they tend tolimit exposure to agricultural finance and to fa-vor clients with established credit histories and

significant collateral. As a result, a relativelysmall number of financial intermediaries have apresence in rural markets and an even smallernumber have significant agricultural lendingportfolios. This limited presence of financialintermediaries in rural areas and the bias againstagricultural lending creates access and segmen-tation problems.

Underdeveloped rural credit markets have seri-ous negative economic and social consequences.Poor access to formal financial services and inparticular credit contributes to persistent pov

Segmentation compounds the inequality andgrowth problems noted above, and also inhibitsinstitutional development, limiting competition,innovation, and efficiency. Institutions tend toidentify small niches and do not concern them-selves with reducing spreads, providing newerservices, or reducing costs because clients havefew alternatives and there are few competitors.Well-capitalized rural financial institutions inLatin America tend to serve only large-scale,agricultural and nonagricultural businesses.Since the large farm/agribusiness segment isvery small to begin with, there is little impetusto innovate or for others intermediaries to con-test this market.

At the other end of the spectrum, microf inance

institutions (MFIs) tend to serve a very smallsegment made up of well-diversified, low-income rural households. MFIs limit loan sizesand term because of their small capital base andthe high degree of unmitigated risk associatedwith agriculture in Latin America. No solid dis-aggregation exists for rural versus urban micro-finance, which is usually defined as loans belowUS$3,000. According to the Microfinance In-formation Exchange (The Mix), there were 762institutions reaching more than 38 million bor-rowers in 2005. 2 A competing database, Micro-credit Summit indicates that there were 92 mil


8/36

CGAP estimates that there are 573 million smallvalue deposit accounts, of which 318 million arepostal savings accounts (Christen, Rosenbergand Jayadeva, 2005).

Yet these smaller institutions cannot grow andexpand services as fast as they would like be-cause they tend to be constrained by capital,technology, and governance structure. Theirlevel of market penetration is increasing but it isstill low. To further exacerbate matters, small-and medium-scale agricultural producers that are

commercially oriented and specialized, tend tobe largely excluded from formal credit eventhough they have demonstrably profitable in-vestment opportunities. They demand amountsthat are larger than what the average microfi-nance institution can provide (US$300-US$3,000), and they lack sufficient collateral,audited financial records, and long credit histo-ries to be deemed creditworthy by large financialinstitution that tend make rural loans greaterthan US$50,000. Because of segmentation, fewinstitutions offer a gamut of financial productssuited to the demands and capabilities of thevarious types of clients or even the same clientover time. Thus, the tiny top of the market isfairly well served, but the middle and bottomsegments are partially served or completely ex-

cluded.

In more developed financial markets a host ofinnovations in credit risk management that have

Latin America because of the lack of supportinginstitutions, but certainly hold insights and les-sons that may help guide improved practices andinnovation in credit risk management.

Policymakers and donor institutions face an effi-ciency and equity challenge in correcting thissuboptimal state of affairs in developing coun-tries. Managers of financial institutions inter-ested in consolidation and growth face the samechallenge; namely, how to better manage risks toreduce costs and improve profit margins. Oper-

ating in rural areas is particularly challengingbecause of the spatial dispersion of clients, hightransaction costs, poor infrastructure, low levelsof education, and low levels of income. How-ever, improvements in risk management tech-niques would go a long way in offsetting theseshortcomings, allowing financial institutions toextend credit services to a hitherto underservedand difficult to reach market segment. Betterrisk management techniques would help finan-cial institutions penetrate rural markets to agreater extent.

The purpose of this report is to review commoncredit risk management techniques used in asample of Latin American financial institutionswith agricultural portfolios, and to identify the

factors that contribute to successful credit riskmanagement as measured by asset quality main-tenance, portfolio growth, and profit margins, inorder to assist the donor organizations and na-


9/36

Review of Typical Risks Faced by Financial Institutionsand Techniques to Address Them

The objective of financial institutions is tomaximize shareholder value by mobilizing de-posits (liabilities) and lending them (assets) tofirms and clients with investment projects. Theinstitution seeks to generate a profit by havinginterest income, fees, and investment or tradingincome exceed the interest paid on deposits, bor-

rowings, and all operating costs. Even if the in-stitution is member-owned or has a philan-thropic motivation, the principle of earning aprofit still applies. Obtaining a positive net in-come is imperative for permanency and sustain-ability. What may differ between a for-profit anda not-for-profit institution is the degree of profitaccumulation and the use of those profits.

Financial institutions face a number of risks inthe pursuit of the aforementioned objective:

credit risks liquidity risks interest rate risks foreign currency risks operational risks (mistakes and fraud com-

mitted by staff) technological risks (power and equipment

failures that lead to data loss)d t i ti i k ( d t f il

fixed interest rate contracts, then the institutioncan become compromised. Similarly, if an insti-tution makes a series of bad loans that cannot berecovered, its viability can be quickly threat-ened. Most of the other risks in and of them-selves usually do not pose fatal threats. Many ofthe other risks would have to be combined in

order to trigger a liquidity crisis. Since the majority of the financial institutions that provideservices to rural areas in Latin America are notdeposit-taking, the focus of the rest of paper willbe on credit risks.4

CREDITWORTHINESS EVALUATIONTECHNIQUES

There are two broad means of evaluating credit-worthiness: appraisal of repayment capacity,and asset-backed lending. The former approachfocuses on investigating the integrity, moralcharacter, management ability, and debt payingcapacity of a potential borrower either throughhuman experts or statistical models, while thelatter focuses on the quality and quantity of as-

sets that can be pledged as collateral and quicklyliquidated in the event of a default.

Repayment Capacity: Human-based Expert


10/36

Figure 1: Credit Analysis Process Flow

Does the borrower have a

repayment strategyWhy does the firm

need to borrow?

Is the risk-reward ratio

acceptable to the

lender?

Source: Adapted from Caoutte et al., 1998

Strategy Review:

Does the firm have a clear sense of direction and

how to get there?

1. Management Analysis: Competence, Integrity

2. Financial Analysis: Balance Sheet and Cash Flow Analysis - projections

3. Sector Analysis: Position in Industry, Price Trends, Competition, Suppliers, Labor Situation, Transport

and Marketing Issues, Importance of Quality and Di fferentiation

4. Qualitative Factors-Environmental Due Dilligence

5. Financial Simulations: Stress Testing, Breakeven Analysis

6. Risk Rating/Legal Opinion

1. Loan Administration: Set up, Funding Schedule, Entry into Data, Management System

2. Negotiations/ Credit Approval

3. Loan Documentation/Closing

cadre of expert credit officers is an expensiveand redundant proposition. Several have to be inconstant training to make allowance for analystswho will leave to work for competitors or inother positions, for those who will have to beforced out due to lack of on the job success and

course of conducting business. To aid the analy-sis, the credit officer usually employs a set ofstandard and specialized industry-specific ratiosthat are used to compare the potential borrowerto industry benchmarks. Some of the mostcommon ratios are listed in Table 1


11/36

Repayment Capacity: Mathematical Models

In the last thirty years, human expert systemshave been slowly displaced in developed coun-try financial markets by credit scoring modelsh i d F l

rarely used in the case of small business and

farm lending. Business scoring models are verydependent on audited financial statements, mar-ket capitalization, and volume of traded shares.In countries where most firms do not keep goodfi i l d d f fi bli l

Table 1: Frequently Used Ratios in Credit Analysis

Category Ratio

Operating Performance Earnings before Interest, Taxes, Depreciation and Amortization(EBTIDA/Sales)Net Income/SalesNet Income/Net WorthNet income/Total AssetsSales/Fixed Assets

Debt Service Coverage

Financial Leverage

EBITDA/Interest Payments >1.5Free cash flow-capital expenditure/interest paymentsFree cash flow- capital expenditures-dividends/interestLong-term debt/capitalization

Long-term debt/tangible net worthTotal liabilities/tangible net worthCurrent liabilities/tangible net worth

Liquidity

Receivables

Current ratioQuick ratioInventory to net salesInventory to net working capitalCurrent debt to inventoryRaw material, work in process, and finished goods as percentages

of total inventoryAging of receivables: 30, 60, 90, 90+ daysAverage collection period

Source: Caoutte et al., 1998* These ratios are commonly used but have to be adapted to specific industries.


12/36

ceivables as collateral to secure loans emergedas alternatives for the land constrained. Com-mercial finance companies pioneered the tech-niques in the 1950s in the United States andcommercial banks entered the market soonthereafter. Asset-backed lending places a pre-mium on valuing and understanding assets andtheir resale markets. Nonetheless, heavy relianceon asset backed financing has three attendantrisks: collateral illiquidity, collateral deprecia-tion, and legal risks. The longer it takes to liqui-date pledged assets, the worse off the lender will

be. Likewise, the lender loses if collateral orpledged inventory suddenly loses market value,deteriorates in storage, or is damaged. Lastly,because asset-based financing results in complexdocumentation, public findings, strict compli-ance with commercial codes, and certain bor-rower impositions, legal errors can prove to bevery costly to the lender. Asset-backed lendingtends to work where there are well-definedproperty rights, uniform commercial codes forall jurisdictions and functioning property regis-tries and court systems. In developing countries,

asset-backed credit evaluation approaches tendto be overly reliant on land as surety. To a lesserextent, liens are also placed on standing crops,livestock, and equipment. The use of inventoryand receivables is rather underdeveloped andrepresents a frontier for developing nations.

COMMON PORTFOLIO CREDIT RISKMANAGEMENT TECHNIQUES

Regardless of the credit evaluation techniquesused to screen and identify good individual

credit risk, a panoply of other strategies existthat are used by lenders to reduce credit risk inthe overall loan portfolio. Table 2 provides a listof the techniques used in financial markets. Allare common except for credit risk insurance andportfolio securitization. The last two are just be-ginning to appear in developing countries, butare more commonplace in Europe (credit insur-ance) and the United States (portfolio securitiza-tion, especially in housing and consumer lend-ing).

Table 2: Strategies for Reducing and Coping with Portfolio Credit Risk

Technique Advantages Disadvantages Implication

GeographicDiversification External shocks (climate,price, natural disasters,

etc.) are not likely to affectthe entire portfolio if therei ti l di ifi ti

If the country is small or theinstitution is capital constrained,it may not be able to apply thisprinciple. It will become vulner-bl t i t i k hi h i

Small financial institu-tions should not beoverly exposed to agri-culture.


13/36

and prepay loan or desert afterloan cycle ends.

Business/Farm Size

Limits

Lender may establish sizethresholds, such as, for

example, that eligiblefarmers must own no lessthan 2 has to apply for aloan. Serves to protectlender from making loansto unviable clients.

Tends to perpetuate financialexclusion.

Government and policy-makers need to make

adjustment in policies tobetter help the excludedpopulation and makethem bankable.

Over

Collateralization

Assures the institution thatenough liquidation valuewill exist for foreclosed

assets.

Excludes poor, low-income cli-ents who are the vast majority ofthe market.

Not a recommendedtechnique if goal is tobetter serve the low- and-

moderate income clients.Joint Liability Loan

Contracts

Assortative matching, peerpressure, and local infor-mation can serve to reducedefault risk.

High transaction costs for bor-rowers due to regular meetingsrequirements and policing ofneighbors.

Has limited applicability.Good for small loansizes but as businessesgrow the demanded loansize may exceed the mu-tual insurance capacityof the group.

Graduated Lending

and TerminationIncentives

In credit-constrained envi-

ronments, clients are eagerto maintain access and willrepay promptly as long asreliable promise of a largerloan exists.

Clients may not be served up to

their repayment capacity, thusopening the possibility of losingprime clients to other competi-tors.

Good technique for a

microfinance institutionto develop a loyal clientbase, but shortly placeshigh pressure on manag-ers to increase loanablefunds. More difficult fora nonregulated, non-deposit taking institu tionto sustain.

Activity/Product

Exclusion Lists

Lender refuses to lend to

certain activities/crops thatare deemed too risky andunprofitable.

The poor may be excluded be-

cause of average returns of ac-tivities (perceived or not) ratherthan calculation based on actualrepayment capacity

May maintain high lev-

els of financial exclusiondepending on the regionor country.


14/36

own funds occurs.Credit Insurance Bank makes clients pur-

chase credit insurance. Inevent of default, bank col-

lects from insurer.

Databases and credit bureausmay not exist to permit insurerto engage in this line of business

in cost-effective manner.Portfolio

Securitization

Lender bundles and sellsloans to a third party.Transfers default risk andimproves liquidity so thatit can continue to lend.Allows lender to developexpertise in analyzingcreditworthiness in one

sector or niche.

Requires well documentedloans and long time series ofperformance data to permit rat-ings and reliable construction offinancial projections.

Requires a well devel-oped secondary market,standardized underwrit-ing practices, and exis-tence of rating compa-nies.


15/36

Summary and Description of Survey Results

The Inter-American Development Bank con-ducted a web-based survey of Latin Americanfinancial institutions with rural portfolios to de-termine perceptions of risks and techniques usedto control, mitigate, and transfer credit risk, aswell as to gauge the financial performance ofsuch institutions. Approximately 225 surveyswere sent out to institutions believed to haverural portfolios, and 42 institutions responded tothe survey in its entirety (see Annex A). The 225institutions were identified by consulting withtrade associations, leading experts, and internaldatabases. All data is self-reported, the samplecannot be considered representative, and suffersfrom nonresponse bias.7 Nonetheless, the surveyprovides some insights and patterns of behavior.

As can be seen in Table 3, the vast majority ofsurveyed institutions are nonregulated and inparticular, nonprofits. The two countries withthe most responding institutions are Bolivia andPeru. In the case of Peru, especialized microfi-nance nondeposit-taking entity (Entidad de De-sarrollo para la Pequea Microempresa, ED-

PYME) has the largest presence in the sample.The agrarian economies of Central America areunderreported.

Table 4 shows that there are marked differencesbetween regulated and nonregulated entities.Regulated institutions are more leveraged, main-tain better asset quality and generate higher re-turns on equity (ROE). As can be expected,regulated entities have significantly more capitaland assets since they can mobilize deposits fromthe public.

Table 5, which presents agricultural credit indi-cators, shows that the agricultural portfolio con-stitutes less than 40 percent of the total loanportfolio. The only exception is the one com-mercial bank reporting, which specializes in ag-ricultural and rural lending and can be consid-ered an outlier. This shows an important prefer-ence for portfolio diversification between agri-

culture and nonagriculture activities.

An analysis of the share of agricultural portfolioversus portfolio-at-risk of all sampled institu-tions yields a positive relationship (Figure 2). Asdiversification diminishes, rural institutions tendto present higher delinquency rates. Amongthese institutions, credit unions tended to havehigher delinquency rates.


16/36

Table 4: General Financial Indicators as of December 2005

Measure No.

Assets(US$)

Equity(US$)

OutstandingLoans (US$)

Delin-quency

(PAR >30

days)

Assets/Equity

ROA(%)

ROE(%)

Regulated Mean 17 112,326,999 8,355,762 59,876,281 2.4 13.4 1.3 15.5Median 17 17,956,254 3,787,223 13,393,088 5.0 5.8 2.5 11.8

Comme rcial Mean 2 673,800,000 27,212,252 289,950,000 0.3 24.8 0.5 9.0Banks Median 2 673,800,000 27,212,252 289,950,000 0.9 20.2 1.0 9.7

NonbankFinancial Mean 11 35,847,882 5,265,965 28,136,743 3.6 6.8 3.9 26.6Institutions Median 11 14,540,379 3,787,223 11,648,147 4.0 4.7 4.0 22.8

Credit Un-ions Mean 4 41,908,070 7,424,457 32,123,150 8.6 5.6 2.1 5.8

Median 4 17,731,786 2,995,952 13,730,540 10.2 7.1 1.4 4.9Nonregu-lated Mean 25 6,291,322 2,783,957 5,314,395 6.1 2.3 3.4 8.0

Median 25 3,281,506 1,780,963 3,043,258 5.0 2.0 4.3 8.3TOTAL Mean 42 49,210,525 5,039,211 27,398,968 2.8 9.8 1.5 13.0

Median 42 6,874,702 2,266,180 5,725,090 5.0 3.9 4.0 10.4

Source: IDB Survey to Rural Financial Institutions, 2006.

Table 5: Agricultural Credit Indicators as of December 2005

AgriculturalPortfolio Number Avg. Loan

Share inTotal Port-

AgriculturalDelinquency PAR>30


17/36

The nature of agricultural lending is quite differ-ent from microenterprise credit, which is gearedprimarily to retail trade (Table 6). In general,agricultural loans tend to be larger, have fewscheduled payments due to sharp seasonal varia-tions in farm-dependent household incomestreams, and interest rates tend to be lower dueto suspected higher interest rate elasticities ofdemand. Agricultural returns tend to be morevolatile than nonagricultural projects; as a result,the debt servicing capacity of farm-dependent

households is more constrained. This means thatrisk-adverse farm household heads would be lessprone to accept higher interest rate contracts,thereby lowering demand for agricultural credit.Again, due to differences in the capital base,regulated entities are more able to grant largerloans than nonregulated entities.

The institutions in the sample reported offeringdifferent financial services to their clients but thegamut tended to be restricted and to favor short-

Figure 2: Agricultural Portfolio Share vs. PAR (>30 days)

0

20

40

60

80

100

0 5 10 15 20 25 30 35PAR>30 days (%)

AgriculturePortfolioShare

(%)



18/36

term working capital loans. For example, allregulated and nonregulated entities offeredworking capital loans, but only 6.3 percent ofthe regulated entities offered leasing comparedto 3.8 percent of the nonregulated; 75 percent ofthe regulated institutions offered fixed invest-ment medium- and long-term loans versus 62percent of the nonregulated ones. Payment ser-vices, transfers, and passbook savings accountswere offered by less than half of the regulatedentities.

As Table 7 shows, the most common type ofloan contract offered was individual loans.Group loans (including associative lending as avariant) were a distant second and village bank-ing was third.

Table 8 indicates that, contrary to conventionalwisdom, financial institutions active in rural ar-eas are more concerned about risks stemmingfrom political interference such as governmentdebt forgiveness programs, passage of usurylaws, and mandatory refinancing of debt thanthey are of weather-related risks. Reduction inprices of agricultural products ranks third in thelist of concerns of rural institutions. It is alsointeresting to note that very few institutions be-lieve that their clients had few investment oppor-

tunities.

Good clients are not scarce. Both nonregulatedand regulated institutions were more or less

ference) about stiff competition, foreign ex-change risks, downturns in the macroeconomicsituation, weak contract enforcement frame-work, lack of liquidity or internal control defi-ciencies. However, there were statistically sig-nificant differences between the two types ofinstitutions. Regulated entities were more con-cerned about the high cost of funds and opera-tional risk than nonregulated institutions.

Nonregulated institutions were more worriedabout the regulatory and legal risks, lack of

guarantees and scarcity of information on re-payment capacity. Regulated, deposit-takinginstitutions were more concerned about interestrate risks due to the need to attract and maintaindeposit accounts.

With regards to credit evaluation methodologies,all respondents used human expert systems.Scoring or mathematical models were rarelyused (1 respondent out of 42). The two key ele-ments to good credit analysis are well-trainedand motivated staff and access to reliable infor-mation.

The majority of the credit officers ranged in agefrom 26 to 35 years old and was male. A littleover 52 percent had both university training and

specialized knowledge of agronomy and agricul-tural sciences and 50 percent had previouslyworked in another financial institution. Sixty-percent of the respondents (17) used perform-


19/36

approved and on-time repayment rates. For 12 of

the 17 respondents using incentives, the com-missions and bonuses amounted to 50 percent ofbase salary. No information was gathered on in-house training.

As part of their credit technology, regulated in-

stitutions tend to have more requirements relatedto the commercial risk of the agricultural activityof their clients. Near half of the sampled regu-lated institutions required clients to have a for-

Table 8: Index of Risk Perceptions

Categories and Subheadings Regulated Nonregulated Total

General External Threats (Average) 0.25 0.25 0.25High Cost of Funding 0.37 0.26 0.31 *Stiff Competition 0.23 0.31 0.28Interest Rate Risk (for deposi t takers) 0.17 0.05 0.10 *Foreign Exchange Risk 0.28 0.15 0.20Regulatory and Legal Risk (difficulties in legalizing guarantees,biased regulatio ns) 0.14 0.52 0.36 *Unfavorable Macroeconomic Situation 0.32 0.24 0.28

External Threats Peculiar to the Agricultural Sector (Average) 0.64 0.71 0.68

Reduction in prices for agricultural products 0.49 0.66 0.59External Risks (bad weather, natural disasters, civil unrest) 0.74 0.69 0.71Political Interference (usury laws, debt forgiveness programs, etc.) 0.68 0.78 0.74

Problems with Clients (Average) 0.27 0.42 0.36 *

Majority of clients lack guarantees 0.42 0.65 0.56 *Majority of clients do not have profitable investment opportunities 0.18 0.20 0.19Weak Contract Enforcement 0.29 0.40 0.36Lack of information on repayment capacity 0.20 0.41 0.33 *

Internal Management (Average) 0.29 0.28 0.28

Lack of Liquidity 0.23 0.29 0.27Operational Risk (fraud committed by personnel, equipment failure) 0.42 0.27 0.33 *Deficient internal controls and weak accounting systems 0.22 0.27 0.25

Source: IDB Survey to Rural Financial Institutions, 2006.Note: Scale is 0-1 with 0 being unimportant or nonexistent, 1 being very important.

* Si nificant difference at least at the 10 ercent level.


20/36

percent)8; (5) repayment incentives to clients(automatic approval for graduated loan, interestrate rebates, etc.) (84 percent); and (6) use ofcredit bureaus (81 percent).9

8 All regulated institutions reported use of provision-ing and risk client classification as part of the tech-niques used to control credit risk.9 The full list includes: (1) portfolio concentrationlimits for the agricultural sector; (2) exposure limit

The information on 22 possible techniques tocontrol credit risk was converted to an indexwith scale 0-7 (0 being no technique used and 7being all techniques used). It became clear thatinstitutions with lower indices of portfolio atrisk relied more on staff incentives than institu-

tions with higher delinquency indices.

They also did not use all the possible measures;but they used selected techniques more effec-

Figure 3: Index of Rural Credit Technology (by components)

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

Information Requirements Guarantees Monitoring Decentralization

Regulated Non-Regulated

Source: IDB Survey to Rural Financial Institutions, 2006.Note: Value for each component of the index goes from 0 to 1, with 0 representing does not use and 1 representing very impor-

tant and used all the time.


21/36

Figure 4: Index of Risk Control Techniques Used: Regulated versus Nonregulated

0.00

0.50

1.00

1.50

2.00

2.50

3.003.50

Regulated Non-Regulated

Strategic Alliances

Third Party Guarantees

Credit Analysis/StaffIncentives

Client Guarantees

Requirements

Lending Policies

Loan Limits


0

0.5

1

1.5

2

2.5

3

3.5

4

0

1

2

3

4

Figure 5: Index of Credit RiskControl

Measures Used: Low v. Hi h PAR

Figure 6: Index of Credit RiskControl Measures Used: Low v.

Hi h E uit


22/36

guarantee funds. Since the majority of potentialrural clients are collateral constrained, the prin-cipal means of controlling credit is through ap-propriate and effective credit analysis processes.

Within the credit appraisal process, the mostimportant elements are information and directmonitoring.

Table 9: Scale Effects Ranked by Total Portfolio Size

Share of Ag-ricultural

Credit (%)

AgriculturalDelinquency PAR>30

days (%)

Total PortfolioDelinquency -

PAR>30 days (%)

ROA

(%)

ROE

(%)

Mean 0.53 4.11 10.60 1.72 0.08Quartile 1 Median 0.61 2.50 3.50 5.0 7.75

Mean 0.25 4.56 5.87 2.57 14.65Quartile 4

Median 0.25 0.25 5.00 2.00 7.00Total Portfolio - Correlation Coefficient -0.308* -0.009 -0.152 0.0238 0.296*

Source: IDB Survey to Rural Financial Institutions, 2006.Note: Due to it significant different size, data from Banco Cooperativo SICREDI was not included in this calculation.(*) Significant of at least 10 percent level (cut-off corresponds to 0.296).


23/36

Summaries of Case Studies in Guatemala and Peru

To complement and deepen the understanding ofhow successful rural financial institutions per-ceive, measure, and control credit risk, four in-depth case studies were conducted in Guatemalaand Peru: Banrural S.A., Fundacin para el De-sarrollo Empresarial y Agrcola (FUNDEA),Caja Munic ipal de Ahorro y Crdito Sullana

(CMAC Sullana), and Entidad de Desarrollopara la Pequea y Micro Empresa Confianza(EDPYME Confianza). The information forthese studies was gathered in the second semes-ter of 2006 principally through field interviewsand review of annual reports, published docu-ments, consultant reports, and information frombanking superintendencies.10

Banrural S.A. is the most profitable commercialbank in Guatemala and the third largest in termsof assets. It has over 300 agencies and conductsthe majority of its operations outside the capital.It grew out of the reform of a failed state agri-cultural bank, maintained a mission to serve ru-ral entrepreneurs and has a mixed capital struc-ture where NGOs, civil associations, and former

employees dominate the board of directors andcontrol the majority of the capital while the gov-ernment is a minority shareholder. FUNDEA isa nonregulated Guatemalan NGO born of the

agricultural producers. Confianza is owned by agroup of local investors and international do-nors.11

As can be seen in Tables 10 and 11, successfulrural financial institutions tended to be well di-versified, profitable, and enjoy high asset quality

both at the level of the general portfolio and atthe level of the agricultural portfolio. Key char-acteristics in explaining their success are thepursuit of diversification strategies, achievementof scale, a well-proven credit evaluation tech-nology, and the retention of skilled and moti-vated staff. Figures 7 through 10 demonstratethat agricultural loans are trending upward for in

all the studied cases for the last available yearsof data, indicating that agricultural lending canbe viable activity.

PERCEPTIONS OF RISK

Consistent with the general findings of the othersurveyed institutions, the four organizations in-cluded in the case study reported price risk faced

by actual and potential clients as a more seriousconcern than climatic risks. In Table 12, pricerisk was mentioned more times in the very im-portant and important risk categories than cli-


24/36

Table 10: General Characteristics of the Four Case Studies as of December 2005Category/Institution

Banrural S.A. CMAC Sullana EPDYME Confianza FUNDEA

Type of Organization Bank Nonbank Nonbank NGORegulated by Banking

Superintendent

Yes Yes Yes No

Ownership Mixed(Private-Public)

Public Private Private

Country Guatemala Peru Peru GuatemalaSphere of Operations Nationwide Northern and

Central Coastarea: Piura,Tumbes, Lam-bayeque, and

capital Lima

Central Highlandsand capital city:Junin, Ucayali, Huan-cayo, Huancavelica,and capital Lima

Central zone, West-ern Highlands, andNorthern zone.

No. of Agencies 323 20 10 19Sectors Served andProducts Offered

Multisectoral(Credit, Savings,Transfers, Pay-ments)

Multisectoral(Credit, Savings)

Multisectoral(Credit)

Predominately rural(Credit)

Outstanding Loans U$673.8 million US$69.5 million US$21.0 million US$6.2 millionNumber of Active Cli -ents

195,822 58,301 26,258 12,213

Share of AgriculturalPortfolio in Total

11.4% 15.2% 14% 39%

Equity US$81.3 million US$15.2 million US$4.9 million US$7.4 millionPortfolio at Risk (>30

days)

0.9% 5.1% 3.5% 0.9%

ROA 2.1% 5.7% 3.7% 12%ROE 38.7% 38.2% 21.3% 14.8%Efficiency (Operating

Costs/ Aver age Gross

Loan Portfolio)

11.4% 11.4% 15.2% 23.5%

Table 11: Agricultural Portfolio as of December 2005


25/36

Figure 8: CMAC Sullana, Peru,Growth in Agricultural Portfolio

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

2001 2002 2003 2004 2005Agricultural Portfolio (US$, 000)

Figure 7: Banrural S.A., Guatemala,Growth in Agricultural Portfolio

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

90,000

100,000

1998 2000 2001 2002 2003 2004 2005 Jun-06Agricultural Portfolio (US$,000)

Figure 9:EDPYME Confianza, Peru,Growth in Agricultural Portfolio

3 000

3,500

Figure 10:Fundea, Guatemala,Growth in Agricultural Portfolio

20 000

25,000


26/36

size of the agricultural portfolio. 13 All of the in-stitutions, however, are quite aware of systemicweather shocks and have developed many of thecurrently used techniques as a result. CMACSullana changed many policies because of the1997/98 El Nio and EDPYME Confianza alsointroduced changes in its technology due to adrought in the central areas of Peru in 2004. Inthe case of Guatemala, both institutions reactedto Hurricane Mitch in 1998 and Tropical StormStan in 2005.

KEYS TO MANAGINGAGRICULTURAL CREDIT RISKS

The four institutions all apply a set of common

trust funds but it accounted for 10 percent offunding. In short, all the institutions have made astrategic commitment to the sector and havelearned how to identify, measure, and managerisks in the sector using largely their own re-sources and applying home-grown solutions.

CRITICAL COMMON ELEMENTS OFTHE CREDIT TECHNOLOGY

Good credit analysis depends essentially on ca-pable staff and on accurate and timely informa-

tion. Other factors such as management informa-tion systems, the application of sophisticatedmathematical techniques, and the availability ofefficient and low-cost communication technol-

Table 12: Perceived Risks

Ranking of

Perceived Risk Banrural S.A. CMAC Sullana

EDPYME

Confianza FUNDEA

Very Important Reduction in Priceof Main Ag. Out-put

Climatic Risk

Important Reduction in Priceof Main Ag. Out -put

Lack of Guaran-tees

Political Interfe r-enceClimatic Risks

Contract Enforce-mentReduction in Priceof Main Output

Somewhat

Important

Climatic Risks Reduction in Priceof Main Ag. Out-

putClimatic Risks

Political Interfe r-ence

Source: Trivelli and Tarazona (2007).


27/36

to reward results. In two cases, performance bo-nuses can sum to 100 percent of base salary. Ascan be seen in Table 13, productivity per loanofficer is extremely high.

Third, copious amounts of information on char-acter, managerial ability, reputation for repay-

ment, and financial viability are gathered andprocessed by the credit analysts.

The analysts rely on credit bureaus, interviews,and personal references. Information is more

important than guarantees. Guarantees are moreformalities and there is no intent to foreclose onthem in the event of a default. Since the legalcosts can be prohibitive to execute land guaran-tees, liens on moveable property and co-signersare preferred.

Fourth, cash flow and sensitivity analysis areused that view the household as the unit of

analysis and not a single line of business or in-vestment project.

Fifth, there is a distinct preference to financehouseholds with diversified streams of income

and that are somewhat insulated from climatic

risks. The key concern is that the household has

sufficient debt service capacity. There are varia-tions in details from institution to institution. Forexample, CMAC Sullana lends primarily to spe-cialized rice farmers. It believes that there is low

price variability due to government protectionand that reliance on irrigated water reduces yieldrisk significantly. However, the minimum sizeof farms tends to be 4 has. This size and multi-cropping assure the lender that the household isviable. In the case of Confianza, there is prefer-ence for households with fragmented plots and avariety of crops.

Sixth, repayment incentives are widely used.The promise of access to a graduated loan andlower transaction costs for repeat loans serves tomotivate clients to avoid strategic defaults.

Seventh, direct monitoring of clients is essential.All four institutions visit clients randomly toreduce moral hazard and to alert upper manage-

ment if the client is likely to default due to ob-served problems.

Because of these critical elements, rural lendingand agricultural lending in particular are ex-tremely labor-intensive and costly.

Table 13: Characteristics of Loan Officers


28/36

PORTFOLIO AND LOAN SIZE LIMITS

As a further means to reduce risk, the four insti-tutions tended to limit exposure to agriculture

ginal and substandard loans have a higher prob-ability of deteriorating into a loss situation andadequate provisioning protects the institution.All four institutions present significantly high

Table 14: Diversification Strategies

Banrural S.A. CMAC Sullana

EDPYME

Confianza Fundea

GeographicDiversification

Operates in the wholecountry.

Operates in fourregions thatcover two dis-tinct agro cli-matic zones

Operates in threeregions, all in oneagro climatic zone.Seeking to expandto Amazon.

Operates in threedistinct agroclimatic zones

SectoralDiversification

Engages in nonfarmmicroenterprisehousing, andconsumer lending

Engages in non-farm microcreditand consumerlending

Engages in nonfarmmicrocredit and con-sumer lending

Engages in non-farm microcreditand rural housinglending

Client IncomeDiversification

Prefers clients withnonfarm sources ofincome

Prefers clientswith nonfarmsources of in-come

Prefers clientswith nonfarmsources of in-come

AgriculturalCommodity

Diversification

Principally livestockfinanced.

Variety of cropsfinanced

Clients must growmore than one cropand ideally must doso in separated plotsat different eleva-

tions or in differentmicroclimates

Variety of cropsfinanced

Index* Not available .66 .60 .62Source: Trivelli and Tarazona (2007).(*) Index weights equally the share of agricultural portfolio in total loan portfolio, the concentration of top three principal cropsin total agricultural portfolio, and the agricultural lending volume of the largest agency as a share of the total agricultural portfo-

lio.


29/36

overall credit technology. CMAC Sullana auto-matically increases provisioning when there areindications that weather-related shocks may

have an adverse impact on the repayment capac-ity of its clients.

Table 15: Portfolio Limits and Loan Size Limits

Banrural S.A. CMAC SullanaEDPYMEConfianza FUNDEA

Explicit Limit onAgricultural Por t-

folio

No Yes, maximum20%

Yes, maximum20%

Yes, 40-45%

Limit on aParticular Crop

No, however doesnot lend for subsis-tence crops or ba-sic grains

No Yes, maximum40%

No, however limitsfinancing of basicgrains.

Maximum Amountfor AgriculturalLoan*

No No No Yes, US$2,800

Limits by Agency No Yes, maximum30% of total port-folio

No Yes, 50-60% of totalportfolio

Source: Trivelli and Tarazona (2007).* Note: Regulated institutions normally cannot lend more than 5 percent of capital to any one party.


30/36

Conclusions and Recommendations

Credit risk management in Latin American ruralfinancial institutions is improving and evolving,but much still needs to be done. Many of theinstitutions surveyed demonstrated success asmeasured by high overall rates of profitability,low delinquency rates in both general and agri-cultural portfolios, and sustained growth rates inagricultural portfolios over time. Nonetheless,the paucity of institutions active in rural areasand expressed desires for better risk manage-ment systems, the relatively small loan sizes,and restricted terms indicate that the situation isless than optimal.

There are four ways to deal with credit riskreduce it, cope with it, transfer it, or retain it.Based on survey results and the four case stud-

ies, the following techniques were identified asthe most important and widely used:

Expert-based, information-intensive credittechnologies (wherein repayment incentivesfor clients and performance incentives forstaff play important roles and informationacts as a virtual substitute for real guaran-tees) are being used to reduce risk.

A number of diversification strategies(geographic, sectoral, commodity) are be-i d i h i k

are publicly-financed loan guarantee funds;however, they are used only modestly (25 per-cent). Historically, guarantee funds have beenplagued with problems of high costs, limitedadditionality, and moral hazard.14 Recent workhas shown that the most successful guaranteefunds in Latin America (in terms of additional-ity) are those in Chile, and that much of the posi-tive impact is due to adequate regulation (Llis-terri et al., 2006). In order to introduce some ofthe other risk transfer instruments more com-monly found in developed financial markets,investments will be needed to reform andstrengthen the insurance industry, capital mar-kets, credit bureaus, commercial codes, securedtransaction frameworks, and information disclo-sure rules.

The implications of using the aforementionedcredit risk management techniques commonlyfound in Latin America are manifold. First, thecredit evaluation technologies commonly usedare very expensive and tend to increase operat-ing costs and interest rates charged because theyare time and labor intensive. Steps need to betaken to dramatically reduce the cost of gather-ing and analyzing data; of securing, perfecting,and executing guarantees; of classifying andmodeling risks; and of monitoring clients. With

t d ti i ti i d li h


31/36

ented institutions can be more readily helped andsupported. Otherwise, the challenge for do-nors/governments and owners of financial insti-tutions is how to rapidly grow and diversify fi-

nancial institutions that started out small with arural vocation and how to attract to rural areaslarger institutions that hitherto were primarilyurban. The majority of rural financial institutionstend to be very small, exhibit many institutionalneeds (access to more low-cost source of funds,inadequate credit technology, better internalcontrols) and are possibly overexposed to agri-

culture. The larger financial institutions that so-cial planners would like to see more active inrural areas are not interested because they per-ceive high risks and can exploit other more prof-itable market segments such as consumer lend-ing to salaried workers.

Third, the agricultural microfinance credit tech-nology reviewed here is essentially an adapta-

tion of urban microcredit technology, but it haslimits. The better-performing institutions seemto adhere to a common set of principles, butthere are slight differences from institution toinstitution as they adapt the principles to suitlocal conditions. For example, the general rule isto give preference to highly diversified house-holds, but if price and yield risk can be con-

trolled, institutions will lend to highly special-ized farm households. The noteworthy differ-ences of the rural adaptation of the urban micro-credit technology are the use of specialized staff

surveys would have to be institutionalized.These changes can be costly and would require anew mindset and way of doing business.

Based on the survey and case study findings, wehave formulated six recommendations for do-nors, governments, and managers of financialinstitutions interested in designing interventionsto improve how rural financial institutions man-age credit risk.

First, donors and governments should identify

and support rural institutions with a minimumscale that would permit easy diversification ofcredit risk and help them to expand and innovateas the preferred or first best option. The secondbest option would be assist those with a clearstrategic commitment to the rural sector andcompetent management to do the following: (i)upgrade credit technologies; (ii) help them de-velop diversification strategies within their reach

(i.e. introduce new credit products, finance awider number of sectors, finance only highlydiversified households); and (iii) use agriculturalportfolio limits by agency and total portfolio asan early warning system to take corrective ac-tions. As the third best option, and in the ab-sence of minimal scale institutions, donors andgovernments should strive to assist smaller insti-

tutions to merge or associate. An effective asso-ciation of smaller institutions can derive benefitsfrom collective action such as fundraising,common training, purchase and installation of


32/36

most difficult to introduce are (i) recognition andvaluing of inventories and accounts receivablesas forms of assets that can be pledged as collat-eral or sold to third parties for cash; (ii) guaran-

tee funds; (iii) credit insurance (death, disability,portfolio); (iv) parametric crop insurance; (v)portfolio securitization; and (vi) derivatives andswaps. Each of the above has preconditions andcountry-by-country assessments would have tobe made. In general, recognition of inventoriesand accounts receivables require reforms inbanking supervision and regulatory frameworks,

commercial codes, and taxes affecting financialtransactions. To improve guarantee fund opera-tions, political interference needs to be mini-mized or eliminated and adequate regulationintroduced. To introduce credit insurance, creditbureaus have to be strengthened, and massivedatabases and probabilistic risk models built. Tointroduce crop insurance, large investments ininformation, training, and modeling are needed.

To introduce portfolio securitization, long dataseries on loan type performance, standard un-derwriting procedures, a sufficient number ofhomogenous loans for bundling, and ratingcompanies are needed. For derivatives andswaps, well-developed legal/regulatory frame-works and capital markets need to be developed.

Third, donors and governments should promoteand support regulated nonbank financial institu-tions. Nonbanks are forced to be more disci-plined (adhere to loan documentation, risk clas-

ture savings (where permitted) besides obtainingcommercial loans), but allowances have to bemade for flexibility and innovation.

Fourth, the role of the state is fundamental inhelping to develop rural financial markets, butdirect political interference at the retail level canretard progress. The preferred role would be forstate-owned second-tier institutions to extendlines of credit and to train staff of rural financeinstitutions. Many of the institutions expressed aneed for more liquidity and access to low-cost

funds. It was also clear that term finance is veryscarce. Most institutions do not offer term fi-nance with the stated reason being fear of mis-matches. Second-tier institutions and interna-tional donors can assist in extending termsthrough a combination of lines of credit andpromotion of savings mobilization.

Fifth, donors and governments should focus on

improving the legal and regulatory framework,especially with regards to improving contractenforcement, an expressed concern of many sur-veyed.

Sixth , donors and governments can assist in thecapture and dissemination of relevant informa-tion that would serve to reduce asymmetries that

contribute to market failures. High quality andfunctioning databases would help to facilitatebetter agricultural marketing, better risk meas-urement, better risk modeling, and the design of


33/36

References

Bank for International Settlements. 2006. Sound Credit Risk Assessment and Valuation for Loans.Basel Committee on Banking Supervision. Basel, Switzerland.

Bhattacharya, H. 1996. Banking Strategy, Credit Appraisal and Lending Decisions: A Risk-ReturnFramework. New Delhi, India: Oxford University Press.

Christen, R. and D. Pearce. 2005. Managing Risks and Designing Products for Agricultural Microfi-nance: Features of an Emerging Model. Occasional Paper No. 11. Consultative Group to As-sist the Poor. Washington, D.C. Available at:http://www.cgap.org/portal/binary/com.epicentric.contentmanagement.servlet.ContentDeliveryServlet/Documents/OccasionalPaper_11.pdf

Christen, R., R. Rosenberg and V. Jayadeva. 2004. Financial Institutions with a Double BottomLine: Implications for the Future of Microfinance. Occasional Paper No.8. ConsultativeGroup to Assist the Poor. Washington, D.C. Available at:http://www.cgap.org/portal/binary/com.epicentric.contentmanagement.servlet.ContentDeliveryServlet/Documents/OccasionalPaper_8.pdf

Caoutte, J. B., E. I. Altman and P. Narayanan. 1998. Managing Credit Risk: The Next Great Finan-cial Challenge. New York: John Wiley and Sons, Inc.

Economic Commission for Latin America and the Caribbean (ECLAC). 2006.Anuario Estadstico deAmrica Latina y el Caribe, 2006. Santiago de Chile.

Gardner, M. and D. Mills. 1991. Managing Financial Institutions: An Asset/Liability Approach.Hindsdale, IL: The Dryden Press.

International Monetary Fund (IMF). 2006. International Financial Statistics. Washington, D. C.Available at: http://www.imf.org/external/data.htm


34/36

Osius, Margaret and B. Putnam.1992. Banking and Financial Risk Management: Training Handbook.Economic Development Institute of the The World Bank. Washington, D.C.

Schreiner, M. and H. Dellien. 2005. Credit Scoring, Banks, and Microfinance: Balancing High-

Tech with High-Touch. Microenterprise Development Review Vol.8 (2). Inter-AmericanDevelopment Bank. Washington, D.C. Available at: http://www.iadb.org/sds/doc/int44D.PDF

Schreiner, M. 2003. Scoring: The Next Breakthorough in Microcredit?. Occasional Paper No.7. Con-sultative Group to Assist the Poor (CGAP). Available at:http://www.microfinance.com/English/Papers/Scoring_Breakthrough.pdf

Trivelli, C. and A. Tarazona. 2007. Riesgo y portafolios agropecuarios: lecciones desde la experien-

cia de instituciones financieras de Amrica Latina. Working paper N151 IEP (Instituto de Es-tudios Peruanos). Available at:http://www.iep.org.pe/textos/DDT/DDT151.pdf

United States Department for Agriculture (USDA). 1997. Credit in Rural America. Agricultural Eco-nomics Report No. (AER749). Washington, D.C.: Rural Economy Division, Economic Re-search Service. Available at: http://www.ers.usda.gov/publications/aer749

Van Greuning, H., and S. Brajovic Bratanovic. 2003. Analyzing and Managing Banking Risk: A

Framework for Assessing Corporate Governance and Financial Risk. The World Bank. Wash-ington, D.C.

Wenner, M. D. 2005. Agricultural Insurance Revisited: New Developments and Perspectives in LatinAmerica and the Caribbean. Sustainable Development Department Best Practice Papers Series,No. RUR-05-02. Inter-American Development Bank. Washington, D.C.

Wenner, M. D. and Francisco J. Proenza. 2006. Rural Finance and the Caribbean: Challenges and

Opportunities. In Promising Practices in Rural Finance: Experiences from Latin America andthe Caribbean. Washington, D.C.: Inter-American Development Bank. Available at:http://www.iadb.org/sds/RUR/publication/publication_210_3301_e.htm


35/36

29

Annex A: List of Participant Institutions in Survey

Institution CountryEquity

(US$ ' 000)

Total Portfolio

(US$ ' 000)

No. of

Loans

Average

Loan (US$)

Agricultural

Portfolio Share

PAR>30

days (%)

PAR>30 days

(%) - Ag.

Portfolio

ROA (%) ROE (%)

Regulated 8,356 59,876 31,672 1,771 52.8% 5.0 4.3 2.5 11.8

FFP Prodem Bolivia 12,164 108,900 68,356 1,593 9% 2.1 2.3 1.9 23.9

Banco Cooperativo Sicredi Brazil 36,902 477,500 33,617 14,204 97% 0.0 0.0 0.4 7.6

Cresol Baser Brazil 23,700 100,900 75,800 1,331 33% 8.0 8.0 2.4 5.4

Codesarrollo Ecuador 3,025 20,691 12,472 1,659 40% 12.4 4.5 0.5 4.5

Banco Procredit El Salvador El Salvador 17,523 102,400 66,617 1,537 1.7 1.6 11.8

Fundacion Microfinanciera Covelo Honduras 3,983 9,975 17,500 570 5% 3.1 4.3 4.0 10.0

Capaz, scl Mexico 6 132 67 1,963 70% 25.0 18.0 -2.2 -36.0

Financiera El Comercio S.A.E.C.A. Paraguay 3,787 13,393 38,518 348 44% 5.0 3.0 6.0 42.0CMAC Sullana Peru 15,215 76,285 74,836 1,019 15% 5.1 2.4 5.3 36.0

EDPYME Edyficar Peru 10,961 54,789 65,202 840 3.7 4.8 24.3

EDPYME Confianza Peru 4,916 22,063 26,256 840 14% 3.5 9.7 4.0 22.8

EDPYME Proempresa S.A. Peru 3,130 11,648 14,536 801 3% 5.0 5.0 4.0 17.0

Cooperativa de ahorro y credito "Santo Cristo de Bagazan" Peru 2,967 6,770 13,067 518 5.0 2.5 10.7

EDPYME Nueva Vision S.A. Peru 1,663 5,400 6,848 789 13% 3.6 0.2 4.5 15.3

EDPYME Efectiva S.A. Peru 1,100 4,000 20,000 200 4.0 15.0 40.0

EDPYME Crear Trujillo Peru 470 2,200 4,100 537 65% 7.3 6.0 0.5 3.4

EDPYME Solidaridad S.A.C. Peru 536 850 630 1,349 60% 14.0 3.0 2.5 5.2

Note: Medians have been used to summarize Portfolio-at-Risks (PAR), Returns on Assets (ROA) and Returns on Equity (ROE).


36/36

30

Annex A: List of Participant Institutions in Survey (continued)

Institution Country Equity Total PortfolioNo. of

Loans

Average

Loan (US$)

Agricultural

Portfolio Share

PAR>30

days (%)

PAR>30 days(%) - Ag.

Portfolio

ROA (%) ROE (%)

Not Regulated 2,784 5,314 7,882 1,079 29.8% 5.0 2.0 4.3 8.3

Fundacion para alternativas de desarrollo - Fades Bolivia 5,584 18,846 37% 5.0 2.0 2.0 7.0Agrocapital Bolivia 10,336 16,334 6,451 2,532 25% 5.0 3.0 2.0 3.0Asociacion Nacional Ecumnica de Desarrollo Bolivia 1,043 12,868 9,996 1,287 67% 15.0 16.0 -0.5 -7.3

Fundacion Diacona FRIF Bolivia 13,079 12,504 26,838 466 2% 1.0 1.0 10.0 11.0Fondo de Desarrollo Comunal - Fondeco Bolivia 2,303 6,050 2,599 2,328 71% 7.0 6.0 1.0 3.0Servicio Financiero Rural Sartawi - Focades Bolivia 277 3,043 2,002 1,520 34% 17.0 14.0 1.3 -0.2

Foncresol Bolivia 1,510 2,029 1,176 1,726 82% 31.0 0.0 12.0 14.0

Corporacion Microempresas de Antioquia Colombia 729 8,750 10,489 834 23% 7.0 1.0 5.0 21.0Corporacion Nario Empresa y Futuro Contactar Colombia 2,229 3,302 3,641 907 14% 6.8 0.1 13.0 18.0

Asociacion ADRI Costa Rica 3,552 14,000 2,500 5,600 5.2 0.0 4.3 20.8Fundacion Mujer Costa Rica 607 684 13.9 1.0 1.3

Fondo Ecuatoriano Populorun Progression - FEPP Ecuador 7,476 2,561 1,465 1,748 73% 10.0 12.0 0.1 0.2

Fundacion Campo El Salvador 2,198 2,142 3,237 662 62% 4.0 3.0 5.0 6.0FUNDEA Guatemala 6,670 6,191 8,185 756 39% 1.0 1.0 7.1 8.3

Asociacion Cooperacion para el Desarrollo Rural del OccidenGuatemala 679 1,697 1,287 1,319 8% 2.8 4.5 10.5 30.9

Asociacion de Desarrollo Integral Cuenca del Lago Atitlan - A Guatemala 344 600 1,131 531 36% 3.0 2.0 5.0 15.0Fundacion Adelante Honduras 40 2,134 7,516 284 80% 2.9 0.4 -21.0 -33.0

Alternativa Solidaria Chiapas, A.C. Mexico 317 890 5,926 150 67% 1.0 0.0 -2.0 -5.0

Caritas del Per Peru 1,611 6,668 24,883 268 6% 1.7 0.0 4.0 24.0Asociacion Benfica Prisma Peru 2,560 4,053 15,222 266 38% 8.0 5.0 -7.0 -9.0

Promujer Peru Peru 2,406 3,416 0.0 10.1 18.0Movimiento Manuela Ramos Peru 1,781 1,817 25,474 71 1.5 2.9 3.5

Finca Per Peru 1,849 1,431 7,279 197 0.5 10.6 11.5Microcrdito para el Desarrollo "La Ch'uspa" - Mide Peru 150 614 4,543 135 4.8 10.2 47.0Instituto para el Desarrollo, Educacion, Salud, y Pacificacion - Peru 270 233 1,563 149 23% 5.1 5.0 8.9 9.5

Note: Medians have been used to summarize Portfolio-at-Risks (PAR), Returns on Assets (ROA) and Returns on Equity (ROE).