The role of financial institutions in value chain finance in the global south

Submitted at Freising, September 30, 2021

Technical University of Munich

The Role of Financial Institutions in

Value Chain Finance in the Global

South

Scientific work to obtain the degree

M.Sc. Agrarmanagement

At the Chair Group Agricultural Production and Resource Economics of the TUM School of

Life Sciences.

Supervisors M.Sc. Roberto Villalba Camacho

Chair Group Agricultural Production and Resource Economics M.Sc.Terese Venus

Chair Group Agricultural Production and Resource Economics

Examiner Prof. Dr. agr. Johannes Sauer

Chair Group Agricultural Production and Resource Economics

Submitted by Moahmed Ali Trabelsi

Matrikelnummer: 03703889

Giggenhausser str. 2985354, Freising [phone]

Declaration of authorship

I, Mohamed Ali Trabelsi, born in Tunis, Tunisia, with matriculation number 03703889; declare that this thesis and the work presented in it are my own. It has been generated as the result of my own original research on the subject «The Role of Financial Institutions in Value Chain Finance in the Global South»

I confirm that:

ü This work was done wholly in candidature for MSc degree thesis fulfillment at the Technical University of Munich.

ü Where I have consulted published work of others, that was always clearly credited.

ü Where I have taken some ideas from other sources, I have mentioned the sources and except these kinds of quotes, the entire work is mine.

Date 30.09.2021 Signature:

i

Table of Contents

Table of Contents ii

List of Figures v

List of Tables vi

List of Abbreviations vii

Acknowledgment viii

Abstract ix

1. Introduction 1

1.1. Problem statement 1

1.2. Objectives 2

1.3. Research questions and hypothesis 2

1.4. Expected results from the research 3

1.5. Organization of the thesis 3

2. Literature Review and Theoretical Background 4

2.1. Agriculture finance 4

2.1.1. Agricultural credit 4

2.1.2. Financial institutions (FIs) in the Agriculture Sector 6

2.1.3. Microfinance 7

2.2. AVCF definition 9

2.3. AVCF Challenges 11

2.4. Competitiveness of agricultural finance 14

2.5. Determinants of agricultural credit 17

2.6. Literature on AVCF 19

2.6.1. Gap of the Literature on AVCF 19

2.6.2. Available literature on AVCF 20

3. Methodology and Data 22

3.1. Description of the study 22

3.2. Building the database 23

3.3. Statistical Analyses 25

3.3.1. Qualitative Analyses: 25

3.3.2. Quantitative Analyses 25

3.3.3. Cluster Analysis 26

3.3.4. Data Types and Variables 27

3.3.5. Other components of the database 30

3.4. Structure of survey 30

ii

3.5. Sample Design 31

3.5.1. Sampling frame 31

3.5.2. Sampling techniques 32

3.5.3. FIs Listing: 33

4. Survey Design and Conceptual Framework 34

4.1. Exploring other surveys 34

4.1.1. Survey with FIs 34

4.1.2. Survey with farmers (World Bank & CGAP) 35

4.1.3. Integrated Financing for Value Chains (WOCCU) 37

4.1.4. Survey on national development bank (World Bank Group) 37

4.2. Credit Scoring for Agricultural Loans 38

4.3. Financial instruments employed by FIs 40

4.4. Survey design for FIs officials 41

4.4.1. General Information 42

4.4.2. Economic information 42

4.4.3. Credit screening, scoring, and monitoring for Agricultural loans 42

4.4.4. Agricultural finance within value chains 43

4.4.5. Financial product & Instrument employed 43

4.5. Overview of the online questionnaire 43

5. Analysis and Results 45

5.1. Descriptive Analysis: 45

5.1.1. Geographic distribution: 45

5.1.2. Distribution by institutional type: 45

5.1.3. Foundation Year: 46

5.1.4. Number of Branches 47

5.1.5. Agricultural loans 47

5.1.6. Gender Equality 48

5.1.7. Digital Solutions 49

5.2. Cluster Analysis 50

5.2.1. Confirm Data: 50

5.2.2. Scale the data 51

5.2.3. Select segmentation variables 51

5.2.4. Define similarity measure: 51

5.2.5. Number of clusters 51

5.2.6. K-means Clustering Method 52

5.2.7. Hierarchical Clustering Method 54

5.2.8. Selected method and number of clusters 55

iii

5.2.9. Extracting Results 57

6. Discussion 62

6.1. An information provider database 62

6.2. Analysis of the clustering analysis 63

6.3. Limitations and further research needs 64

7. Conclusion and Recommendations 67

7.1. Conclusion 67

7.2. Recommendations 68

References 70

Annex A I

Annex B: Database II

Annex C: Survey for FIs Officials IX

Annex D: R Script XVI

iv

v

List of Figures

Figure 1: Sources of agriculture credit 5

Figure 2: Components of direct and indirect agriculture credit 6

Figure 3: Development Process through Micro-finance 7

Figure 4: Overview of value chain finance Triangle 10

Figure 5: Considered factors to reduce TC and risk management in agricultural finance 15

Figure 6: Geographical location of the financial institutions 22

Figure 7: Geographical distribution of financial institutions 32

Figure 8: Listing criteria of financial institutions in the final database 33

Figure 9: Financial institutions Questionnaire components 41

Figure 10: List of instruments enquired during the survey 43

Figure 11: Questionnaire Framework 44

Figure 12: Determination of the optimum number of clusters 52

Figure 13: Grouping Data scaled in different clusters 53

Figure 14: Clusters Visualization 53

Figure 15: Hierarchical Clustering 54

Figure 16: Validation of the number of clusters 57

vi

List of Tables

Table 1: Distribution of MFIs by institutional type 8

Table 2: The challenges of the agriculture finance according to the literature review 12

Table 3: Determinants of credit Access 19

Table 4: Number of FIs mentioned in the literature review 21

Table 5: Composition of database 24

Table 6: Most popular Qualitative Analysis method 25

Table 7: Quantitative statistic types 26

Table 8: Data types 27

Table 9: Continent Attributes 27

Table 10: Institutional Type Attributes 27

Table 11: Agricultural loans Attributes 28

Table 12: Gender Attributes 28

Table 13: Digital Solutions Attributes 29

Table 14: financial Institutions General Information 34

Table 15: Specific loan features 35

Table 16: focal points of the World Bank Survey 36

Table 17: Lending decision variables 39

Table 18: Loan accreditation Characteristics 40

Table 19: AVCF Instrument 40

Table 20: Classification of financial institutions by continent 45

Table 21: Classification of financial institutions by institutional type 46

Table 22: classification of financial institutions by foundation year 47

Table 23: Classification of financial institutions according to the number of branches 47

Table 24: Percentage of credit offered by type of financial institution 48

Table 25: Percentage of gender equality program offered by type of financial institution 49

Table 26: Percentage of digital solutions offered by type of financial institution 49

Table 27: Descriptive statistics of the dataset 50

Table 28: Cluster membership IDs using K means method 54

Table 29: Cluster membership IDs using Hierarchical method 55

Table 30: Nomination of the FIs groups 57

Table 31: Cluster's characteristics 60

List of Abbreviations

vii

ADB Asian Development Bank

AFD Agence française de développement

AfDB African development Bank

AFRACA African Rural and Agricultural Credit Association

AL1 Farmer credit

AL2 Agribusiness Credit

AOI Agriculture orientation index

AVCF Agricultural value chain finance

CB commercial banks

CGAP Consultative Group to Assist the Poor

DBs Development Banks

DS1 Online Banking

DS2 E-products Email and SMS Banking

DS3 Online loan application

FI Financial institution

G1 Credit facility for women

G2 Career development opportunities to female staff

G3 Gender Programmes

GS Global South

IDFC International Development Finance Club

IFAD International Fund for Agricultural Development

IFC International Finance Corporation

IFI International financial institutions

IIRR International Institute of Rural Reconstruction

INSE Institute of New Structural Economics at Peking University

ISF advisory group

KIT Royal Tropical Institute

MBFI membership-based financial institutions

MFCs microfinance Companies

TC transaction costs

VC value chain

VCF value chain finance

WFDFI World Federation of Development Financing Institutions

WBG World Bank Group

WOCCU World council of credit unions

Acknowledgment

VIII

The first thing I want to say is how grateful I am to God and my father for the opportunity to study at the Technical University of Munich. I would also like to extend my deepest gratitude to my sister Emna for her support throughout the entire thesis process. Also, I want to thank my girlfriend Myriam, as well as my friends Anas, Christian, Cyrine, Dali, Mourad, Rached, Ramzi, Sabrina, Safa, Youssef, Wajih, Werner, and Zeineb for their continued encouragement.

My thanks go out to Villalba Camacho Roberto and Venus Terese for their helpful guidance and valuable comments and corrections during this work. This work cannot be done without them. My strong gratitude to Susanne Minges and Papaja-Hülsbergen Susanne for their continuous support during my studies at TUM.

The internship opportunity with Agribusiness Facility for Africa (ABF) and Green Innovation Centres for the Agriculture and Food Sector (GIC) at the GIZ was a big milestone in my career development. It was a great chance for learning and applying my knowledge and fresh skills in a real working setting. I will strive to use everything I learned in the best possible way. Through this internship, I met wonderful people and professionals who helped me develop my experience. I would like to express my deepest thanks to Dr. Annemarie Mathess, Carsten Schüttel, Wahid Marouani, and Melanie Hinderer for their guidance and allowing me to participate in their projects which helped me expand our knowledge on various topics.

ix

Abstract

Small-scale farmers and agribusinesses in the Global South still face many barriers to access credit, despite the efforts of development agencies, facilitator, and even financial institutions. An agricultural gap persists that limits sector potential. The current master's thesis examines the ways in which financial institutions make credit easier to obtain for smallholder farmers and value chain players. The study uses a unique database of 347 financial institutions in 106 countries from Africa, Asia, South America, and Oceania as well as international institutions. It primarily contains cooperatives, commercial banks, NGOs, microfinance institutions, and agricultural banks. The database is constructed through a snowball effect process using literature sources, online search, and open-source bank platforms. This database contains several details about these institutions including their institutional type, agricultural loans, gender equality, and digital solutions.

The number of financial institutions was reduced to 144 for the statistical analysis due to the lack of available data for several financial institutions. Then an analysis of clusters was conducted to answer the research question and determine patterns, similarities, and differences among the selected financial institutions. Five clusters were identified. It emerges from the study that financial institutions deliver customized and enhanced rural financial services in high demand and in line with gender issues, as proved in clusters 1, 2, and 3. Moreover, the youngest group of institutions is cluster 4, which has the most digital solutions to offer. Cluster five, which contains individuals primarily using traditional banking methods, has the lowest level of financial services. This study has addressed the research question in terms of credit provision, gender promotion, and digital solutions, as well as identified the kind of similarities and differences between financial institutions. Several recommendations are made in this study, including the need to encourage women and provide digital solutions to ease the lending process for small-scale farmers and value chain actors.

Keywords: Agricultural Value Chain Finance, financial institutions, cluster analysis, financial products, Global South

x

Zusammenfassung

Kleinbauern und Agrarunternehmen im globalen Süden sehen sich trotz der Bemühungen von Entwicklungsorganisationen, Vermittlern, Maklern und sogar Finanzinstituten immer noch vielen Hindernissen beim Zugang zu Krediten gegenüber. Es besteht weiterhin eine Lücke in der Landwirtschaft, die das Potenzial des Sektors einschränkt. In der vorliegenden Masterarbeit wird untersucht, wie Finanzinstitute Kleinbauern und Akteuren der Wertschöpfungskette den Zugang zu Krediten erleichtern können. Die Studie stützt sich auf eine einzigartige Datenbank von Finanzinstituten in 106 Ländern Afrikas, Asiens, Südamerikas und Ozeaniens sowie von internationalen Institutionen. Sie enthält vor allem Verbände, Geschäftsbanken, NGOs, Mikrofinanzinstitute und Landwirtschaftsbanken. Die Datenbank wurde in einem Schneeballeffekt-Verfahren unter Verwendung von Literaturquellen, Internetrecherchen und Open-Source-Bankenplattformen erstellt. Die Datenbank enthält verschiedene Details über diese Institutionen, darunter ihre institutionelle Einrichtung, Agrarkreditangebote, Frauenförderung und digitale Lösungen.

Die Zahl der Finanzinstitute wurde für die statistische Analyse auf 144 reduziert, da für mehrere Finanzinstitute keine Daten vorlagen. Anschließend wurde eine Clusteranalyse durchgeführt, um die Forschungsfrage zu beantworten und Muster, Ähnlichkeiten und Unterschiede zwischen den ausgewählten Finanzinstituten zu ermitteln. Es wurden fünf Cluster identifiziert. Aus der Studie geht hervor, dass die Finanzinstitute maßgeschneiderte und verbesserte Finanzdienstleistungen für den ländlichen Raum anbieten, welche auf geschlechtsspezifische Aspekte Wert legen, wie in den Clustern 1, 2 und 3 nachgewiesen wurde. Darüber hinaus ist die jüngste Gruppe von Instituten in Cluster 4 zu finden, die die meisten digitalen Lösungen zu bieten hat. In Cluster 5, in dem sich Finanzinstitute befinden, die hauptsächlich traditionelle Bankmethoden nutzen, ist das Angebot an Dienstleistungen am geringsten. In dieser Studie wurde die Forschungsfrage in Bezug auf die Kreditvergabe, die Förderung der Geschlechtergleichstellung und digitale Lösungen beantwortet und die Gemeinsamkeiten und Unterschiede zwischen den Finanzinstituten ermittelt. In dieser Masterarbeit werden mehrere Empfehlungen ausgesprochen, darunter die Notwendigkeit, Frauen zu fördern und digitale Lösungen anzubieten, um die Kreditvergabe für Kleinbauern und Akteure der Wertschöpfungskette zu erleichtern.

Schlüsselwörter: Finanzierung der landwirtschaftlichen Wertschöpfungskette, Finanzinstitute, Clusteranalyse, Finanzprodukte, Globaler Süden.

xi

Résumé

Les petits exploitants agricoles et les SME des pays du Sud sont toujours confrontés à de nombreux obstacles pour accéder au crédit, malgré les efforts des agences de développement, les facilitateurs, les gouvernements et même des institutions financières. Un écart financière agricole persiste qui limite le potentiel du secteur. La présente thèse de mémoire examine les moyens par lesquels les institutions financières facilitent l'obtention de crédits pour les petits exploitants agricoles et les acteurs de la chaîne de valeur. L'étude utilise une base de données unique de 347 institutions financières dans 106 pays d'Afrique, d'Asie, d'Amérique du Sud et d'Océanie ainsi que des institutions internationales. Elle contient principalement des coopératives, des banques commerciales, des ONGs, des institutions de microfinance et des banques agricoles. La base de données est construite par un processus d'effet boule de neige en utilisant des sources documentaires, des recherches en ligne et des plateformes bancaires à code source ouvert. Cette base de données contient plusieurs détails sur ces institutions, notamment leur type d'institution, les prêts agricoles disponible, la promotion de l'égalité des sexes et les solutions digitales offertes par l'institut.

Le nombre d'institutions financières a été réduit à 144 pour l'analyse statistique en raison du manque de données disponibles pour plusieurs institutions financières. Ensuite, une analyse typologique a été menée pour répondre à la question de recherche et déterminer les modèles, les similitudes et les différences entre les institutions financières sélectionnées. Cinq regroupements ont été identifiés. Il ressort de l'étude que les institutions financières fournissent des services financiers ruraux personnalisés et améliorés, très demandés et conformes aux questions d'genre, comme le prouvent les regroupement 1, 2 et 3. En outre, le groupe d'institutions le plus jeune est l'amas 4, qui a le plus de solutions numériques à offrir. La grappe 5, qui contient des individus utilisant principalement des méthodes bancaires traditionnelles, a le niveau le plus bas de services financiers. Cette étude a répondu à la question de recherche en termes d'offre de crédit, de promotion du genre et d'offre de solutions numériques, et a identifié les similitudes et les différences entre les institutions financières. Plusieurs recommandations sont formulées dans cette étude, notamment la nécessité d'encourager les femmes et de fournir des solutions numériques pour faciliter le processus de prêt pour les petits agriculteurs et les acteurs de la chaîne de valeur.

Mots-clés : Financement de la chaîne de valeur agricole, institutions financières, analyse typologique, produits et service financiers, Sud global.

ÕÎáãáÇ

ÒÒÒÒØ ÏÒÒÒÒíÏ áÛ æÒÒÒÒåÌÛæí æÒÒÒÒäÌáÛ ÇæÏ ÒÒÒÒ ÒÒÒÒ æÉØáÛ æ Ç ÒÒÒÒ áÛ ÒÒÒÒÔáÛæ íÒÒÒÒÍáÇáÛ ÒÒÒÒ ÇÛÕÒÒÒÒí Çá

ÏæÒÒÒÒåÌ ÒÒÒÒØ ÒÒÒÒáÛ ÒÒÒÒ åÉ ÒÒÒÒÔä ÒÒÒÒÍáÇáÛ ÇÒÒÒÒíæØÉáÛ ÒÒÒÒ ÇæÒÒÒÒ ÍáÛ æÏ ÇæÒÒÒÒÍÉ ÒÒÒÒÉáÛ ÕÌÛæÒÒÒÒÍáÛ

æÒÒÒÒÌáÛ ÒÒÒÒå ÏæÒÒÒÒÌæ Û ØÉÒÒÒÒ Û ÒÒÒÒíá ØáÛ ÒÒÒÒ áØáÛ ÒÒÒÒÉÍæ ì ÒÒÒÒ æáÛæ ÒÒÒÒØæ ÍáÛæ ÒÒÒÒíØäÉáÛ Çá ÒÒÒÒ æ

í ÉÒÒÒ Ì ØáÛ ÒÒÒÍæ Í Ñ ÏÒÒÒÉ ÒÒÒÍáÇáÛ ÒÒÒ áÛ ÒÒÒä ØÇ ÒÒÒØ ÏÒÒÒÍáÛ ÒÒÒ ÒÒÒí Ç ÒÒÒÔ å ÒÒÒ É ÒÒÒí Û Õ áÛ

ÍÒÒÒ Í æ í ÛÕÒÒÒØ á ÇÒÒÒíæØÉáÛ ÒÒÒ ÇæÒÒÒ ÍáÛ ÒÒÒíá ØáÛ ÒÒÒ áØáÛ ÒÒÒå ÇåÒÒÒÒ É ÒÒÒÉáÛ Ó ÒÒÒ áÛ ÒÒÒíá ÍáÛ

ÒÒÒä í ÏÒÒÒ Ç ÒÒÒ Û ÏáÛ ÏÇÉÒÒÒ É ÒÒÒÉä Û ÒÒÒ ÒÒÒ í ÒÒÒáÛ ÇÒÒÒ æ ÒÒÒ æÉØáÛ æ Ç ÒÒÒ áÛ ÒÒÒ áØáÛ

ÒÒÒÖ íÒÒÒ æä íÇæÍæ ÒÒÒí æäÌáÛ ÒÒÒ í ØÍæ íÒÒÒ æ ÒÒÒí í Ç ÒÒÒØ ÒÒÒáæÏ 106 ÒÒÒØ ÒÒÒíá ØáÛ ÒÒÒ áØ á ÏÒÒÒí

ÒÒÒí ÌÉáÛ æÒÒÒä áÛæ ÒÒÒíäæ ÉáÛ ÒÒÒ ÒÒÒ Í Ç ÒÒÒÔ ÏÒÒÒ áÛ ÒÒÒå æÒÒÒÉÍÉ ÒÒÒíáæÏáÛ ÇÒÒÒíæØÉáÛ ÒÒÒ á Ø ÒÒÒáÇ

ÏÒÒÒ áÛ ÒÒÒå ì ÒÒÒÔäÇ É ÒÒÒíÍáÇáÛ æÒÒÒä áÛæ Ç ÒÒÒ áÛ ÇÒÒÒíæØÉáÛ ÒÒÒ áØæ ÒÒÒíØæ ÍáÛ ÒÒÒí ÒÒÒØÙäØáÛæ

ÏÒÒÒÒ ØáÛ ÒÒÒÒÍæÉØ æÒÒÒÒä áÛ ÒÒÒÒ äØæ ÒÒÒÒä Éä Û ÒÒÒÒ æÒÒÒÒÍ áÛæ ÒÒÒÒí Ï Û Ï ÒÒÒÒ Ø ÛÏÇÉÒÒÒÒ ÒÒÒÒä í á

ÒÒÒå æä ÒÒÒá ÒÒÒ ÒÒÒØ ÒÒÒ áØáÛ ÒÒÒå ÇæÒÒÒÍ ÇíÒÒÒ ÉáÛ ÒÒÒØ ÏÒÒÒíÏ áÛ ÒÒÒ ÒÒÒå ÒÒÒä í áÛ ÏÒÒÒ Ç æÒÒÒÉÍÉ

ì Í á æÉØáÛ íØÇ áÛ Çæ ÍáÛæ í äÌáÛ í Ûæ ØáÛ ÌØÛ æ í Û ÕáÛ Öæ áÛæ áØáÛ

ÏÒÒÒ ÒÒÒ ÒÒÒ Í Û ÇÒÒÒí ÍÉ á ÒÒÒ áØ 144 ÒÒÒáÇ 347 ÒÒÒØ ÒÒÒíá ØáÛ ÒÒÒ áØáÛ ÏÏÒÒÒ ÇíÒÒÒ É ÒÒÒÉ ÒÒÒØ

ÒÒÒ ÒÒÒ ÌáÅá ÏæÒÒÒ ä áÛ ÇÒÒÒí ÍÉáÛ ìÛ ÒÒÒÌÇ ÒÒÒÉ ÒÒÒ ÒÒÒíá ØáÛ ÒÒÒ áØáÛ ÒÒÒØ ÏÒÒÒíÏ á ÒÒÒ ÇáÛ ÒÒÒä í áÛ æÒÒÒÉ

ÏÒÒÒíÏÍÉ ÒÒÒÉ ÒÒÒÉÇØáÛ ÒÒÒíá ØáÛ ÒÒÒ áØáÛ íÒÒÒ ÒÒÒáÇÉÇáÇÛæ å ÒÒÒÔÉáÛæ ÒÒÒØä Û ÏÒÒÒíÏÍÉæ æÒÒÒÍ áÛ ÒÒÒ Í

ÒÒÒÒíí ÒÒÒÒíá Ø ØÏÒÒÒÒÇ ÏÒÒÒÒ É ÒÒÒÒÉáÛ ÒÒÒÒíá ØáÛ ÒÒÒÒ áØáÛ Í ÒÒÒÒ Û ÏáÛ ÒÒÒÒØ ÒÒÒÒÖÉí ÒÒÒÒ æØÌØ ÑÒÒÒÒØÇ

2 æ 1 ÒÒÒ æØÌØáÛ ÒÒÒ ÒÒÒ ÒÒÒØ íÒÒÒ äÌáÛ íÒÒÒ Ûæ ÒÒÒ ØáÛ ÍÏÒÒÒ Øá ÒÒÒØ Ï ÒÒÒ Û ÒÒÒå ÕÕÒÒÒ Øæ ÒÒÒ ÇØ

ÒÒÒÉáÛ 4 ÒÒÒ æØÌØáÛ ÒÒÒå ÒÒÒ áÛ ÒÒÒ æÉØ ÒÒÒíÍ ä ÒÒÒØ ÒÒÒ æØÌØ ÒÒÒ Í äÒÒÒ ÒÒÒá ÒÒÒ æáÇ ÒÒÒ æ 3æ

í ÒÒÒáÛ ÏÛ ÒÒÒ Û ÒÒÒ æÒÒÒÉÍÉ ÒÒÒÉáÛ ÒÒÒ Ø ÇáÛ ÒÒÒ æØÌØáÛ åØÏÒÒÒ É ÒÒÒÉáÛ ÒÒÒíØÇ áÛ ÇæÒÒÒ ÍáÛ ÒÒÒÙ Ø ÒÒÒÖÉ

ÒÒÒíá ØáÛ ØÏÒÒÒÇáÛ ÒÒÒØ ÇæÉÒÒÒ Ø ÒÒÒäÏÍ åíÏÒÒÒá ÒÒÒíÏí ÉáÛ í ÒÒÒ ØáÛ íá ÒÒÒ Û Çæ Û ÒÒÒ ØáÛ ÒÒÒ æØÏÇÉÒÒÒ í

ÕÒÒÒíÕ Éæ ÒÒÒÍáÇáÛ ÇÒÒÒíæØÉáÛ íæÒÒÒÉ æÒÒÒíÍ ÒÒÒØ æÒÒÒÍ áÛ á ÒÒÒ Ø ÒÒÒ Û ÏáÛ ÒÒÒå ÒÒÒáæ äÉ ÒÒÒíÍáÇáÛ ÒÒÒíá ØáÛ

íÒÒÒÒ áÇÉÒÒÒÒÇáÇ Ûæ ÒÒÒÒÔÉáÛ ÒÒÒÒÌæÍ æÒÒÒÒä ÏÏÒÒÒÒÍ á ÒÒÒÒ æ ÒÒÒÒíØÇ áÛ ÇæÒÒÒÒ ÍáÛæ íÒÒÒÒ äÌáÛ íÒÒÒÒ Ûæ ÒÒÒÒ ØáÛ

ÒÒÒ Û ÏáÛ ÒÒÒå ÒÒÒ íÒÒÒ æÉáÛ ÒÒÒØ ÏÒÒÒíÏ áÛ íÏÒÒÒ É ÒÒÒÉ ÒÒÒÍáÇáÛ ÇÒÒÒíæØÉáÛ ÒÒÒ ÒÒÒ áÛ ÒÒÒíá ØáÛ ÒÒÒ áØáÛ

ÒÒÒ á ÖÛ ÒÒÒÇ Û ÒÒÒí Ø ÇíåÒÒÒ Éá ÒÒÒíØÇ ÇæÒÒÒ Í íÏÒÒÒ Éæ ì ÒÒÒ äáÛ ííÌÒÒÒÔÉ ÒÒÒáÇ ÒÒÒÌ ÍáÛ ÒÒÒá ÒÒÒ ÒÒÒØ

æäÌáÛ ÇæÏ Éä Û íØå ØáÛ Ç æ íÍáÇáÛ

ÇæÏ ÒÒÒÒíá ØáÛ ÇæÒÒÒÒ ÍáÛ ÏæÒÒÒÒ ä áÛ ÇÒÒÒÒí ÍÉáÛ ÒÒÒÒíá ØáÛ ÒÒÒÒ áØáÛ ÒÒÒÒÍáÇáÛ ÇÒÒÒÒíæØÉáÛ : Í ÒÒÒÒÉØáÛ ÒÒÒÒØ áÛ

xii

æäÌáÛ

1

1. Introduction

1.1. Problem statement

Agriculture in developing countries is undergoing major changes, including globalization and the transition from traditional low production agriculture to modern high production agriculture. The result of this process of profound changes has important consequences on poverty, risk management and agricultural smallholders' income (Abid, Jie, Aslam, Batool, & Lili, 2020). Smallholders face severe problems resulting from the specificity of the production cycle. They have also to deal with climatic factors such as extreme weather shocks and biological factors like insect pests, crop, and livestock diseases (Fries & Akin, 2004). These production risks are linked with price and market risks. Therefore, the variability of production provokes high food price instability (Antonaci, Demeke, & Vezzani, 2014). Due to this high risk, financial institutions are less interested in financing the agricultural sector because of low profit and low collateral (Herliana, Sutardi, Aina, Himmatul, & Lawiyah, 2018). Moreover, Financial Institutions (FIs) consider micro-entrepreneurs as "non-bankable», or not creditworthy because they have no previous credit history or guarantee to offer (Yunus M. , 2007). On the other side, farmers often face multiple challenges to access the finance they require, the outcome is thus a financing gap that limits the potential of agriculture (UNCTAD, 2004). This financing gap which exists in the agricultural sector is estimated at about $170 billion per year (ISF Advisors and Mastercard Foundation, 2019). Development agencies, research institutes and donors have centered their efforts on developing new approaches that allow different stakeholders, such as agribusiness, and financial institutions to address this gap. The aim is to provide innovative financial services to producers, processors and traders as well as develop an economic and financial environment (IFAD and CPI, 2020).

Among these approaches, we can find Agricultural Value Chain Finance (AVCF) which refers to leveraging the a value chain's connections and social capital to improve financial flows and reduce the risks in the chain (Miller & Jones, 2010). Whereas many of the value chain finance transactions, instruments, and processes are not new (Robert, Chalmers, & Grover, 2012),what is new is how AVCF is used by FIs and rural producers. What is also innovative is the variation of the application and the different organizations that offer finance in different innovative ways, as well as the diversification, the intensification, and combination of mechanisms (Miller & Jones, 2010). It also means linking financial institutions to the value chain, providing financial services to support the flow of products, and building on the relationships established at the chain level. This type of financing offers alternatives to traditional requirements (KIT and IIRR, 2010). This allows all value chain participants to benefit from it without collateral requirements (Cuevas & Pagura, 2016). AVCF differs from other types

2

of financing in that it expands financing opportunities for agriculture and improves repayment efficiency among chain participants. It is not only that the nature of the funding is different, but also the motivations. Nyoro (2007) mentions that `value chain actors are driven more by the desire to expand markets than by the profitability of the finance' (Nyoro, 2007). The solutions offered by AVCF can help to build a value chain, mitigate barriers, or improve value chain operations, thereby increasing the competitiveness of the chain (KIT and IIRR, 2010). The challenges that AVCF can face are legal systems that enforce contracts and provide some type of ownership, lack of bank penetration and institutions offering loans for investment in rural areas, high transactions cost, lack of knowledge and developed infrastructure (Zander R. , 2016).

1.2. Objectives

Recent work has focused on evaluating the access to finance at the farmers' level (Gamage, 2013), however, there is limited evidence on the role of Financing Institutions (Meyer R. L., 2002), in particular, in new approaches such as Agricultural Value Chain Finance. This master's thesis aims to build a database of financial institutions that fund agriculture in the Global South. A number of different financial institutions are included in the database, including a range of institutional types and banking experience, as well as the services offered by each institution. Following that, a descriptive analysis of the data from these institutions will describe the basic features of the data. This analysis will provide simple summaries of data and draw conclusions from it. A later study can use this database to conduct the online questionnaire with these FIs. This will enable research staff to determine the methods that financial institutions use about credit screening, scoring, and agricultural value chain financing. A part of this thesis involves analyzing the existing questionnaire and preparing the basis for the design of the questionnaires [Annex C]. From this database, a cluster analysis using R will be able to draw conclusions about lending to farmers and credit for value chain participants. The aim of this study is to provide robust evidence regarding the similarities and differences between financial institutions when it comes to offering rural services to their clients.

1.3. Research questions and hypothesis

The present study will focus on the role of financial institutions in implementing value chain finance in the Global South. This study will address the following questions:

1) What are the key underlying characteristics of credit provision of different types of financial institutions in the Global South?

2) What is the extent to which financial institutions promote gender issues and offer digital solutions?

3

3) What kind similarities and differences can be observed between financial institutions? As a result, the present work highlights the subsequent main hypotheses H1, H2 and H3:

H1: Various trends can be seen on the basis of the variables concerning the provision of credit by different types of financial institutions in the Global South.

H2: Only a few financial institutions deal with gender issues and offer digital solutions

H3: Financial institutions show several similarities and differences in credit provisions, gender programs, and digital solutions.

1.4. Expected results from the research

The expected outcomes of this study are:

i. A database of financial institutions which fund agriculture in the Global South.

ii. Data-driven evaluation of financial institutions' services in the Global South

iii. An agenda for agricultural finance policy recommendations

1.5. Organization of the thesis

Throughout this study, six sections are discussed. The following part is a literature review which covers theoretical perspectives about agriculture finance, AVCF definitions and challenges, and a review of available papers, as well as agricultural credit determinants. In the third section, we describe the study, the way the database was built, and the statistical methods used. The fourth section focuses on the design of the survey. In part five, we examine the results of our descriptive and cluster analyses. Lastly, the fifth part summarizes the findings, discusses them, and makes policy recommendations. The Annex «C» contains the survey.

4

2. Literature Review and Theoretical Background

2.1. Agriculture finance

2.1.1. Agricultural credit

Several interventions are needed in the form of financial institutions and instruments at the value chains (VCs) level to improve financing in the agricultural sector. This interest has been renewed following the economic crisis and the increase in food prices in 2008 and 2011 (Arias, 2019). In most developing countries, the level of financing and public expenditure on agriculture remains very low. This reflects the low percentage of the share of agricultural credit of total credit (Piñeiro, 2019). Moreover, according to the agriculture orientation index (AOI) for credit, agriculture financing is still also low and is just 0.4 in developing countries, while in developed countries it is 1.95 (Arias, 2019).

According to Adams (1994), rural producers need access to financing at the right time and this needs to be stable and reliable for more than a few cropping seasons. This is designed to improve the production and marketing process as well as to have access to input, new technologies and limited resources (Zander R. , 1994). While rural credit is a powerful instrument for poverty alleviation (Ololade & Olagunju, 2013), supporting the agricultural sector is always complicated for FIs because access to information is very expensive and difficult. Moreover, the soft skills in lending to small-scale farmers are not well developed (Zander R. , 2016).

Many studies on the agricultural credit in developing economies have shown that agricultural lending is necessary to improve productivity in the agricultural sector (Sriram, 2007; Das, Senapati, & John, 2009). Other studies have found that without external financing, small-scale farmers cannot even continue their business, and this is proved in the history and debts of people working in agriculture (Gowhar, Ganie, & Padder, 2013). Agricultural credit is therefore a necessary element in meeting the need for investment and bridging thegap between the farmer's income and the expenditure in the field (Khan, Shafi, & Shah, 2011). Additionally, Agriculture credit plays a key role in the modernization of agriculture by removing financial constraints and accelerating the adoption of new technologies (World Bank, 1975).

In this context, Singh et al. (2001) announced that most farm households face a lack of funds on their side. To meet their credit requirements, both formal and non-formal financing is available in a developing economy. Pradhan (2013) suggests that farmers' need for credit increased spontaneously after the Green Revolution. This was the period when institutional sources of credit were considered major players. This was the time when subsistence crops were replaced by cash crops. These credit sources were classified in the following figure into

three groups by Yadav and Sharma (2015) following their intensive literature review on agricultural credit.

Agriculture credit

Non-
institutional
sources

Semi-
institutional
Sources

Institutional Sources

5

Figure 1: Sources of agriculture credit

Source: Yadav & Sharma (2015)

Figure 1 shows the main sources of credit that are available to rural producers. Credit from institutional sources includes credit from the creation of institutional framework with banks and institutions including specific organizations established for agricultural development, commercial banks, cooperative banks, and regional rural institutions. Non-institutional sources cover credit from the unorganized sector such as friends, relatives, landlords, entrepreneurs who are not part of the institutional set-up (Ijioma & Osondu, 2015). Halfway between the institutional and non-institutional agencies is the semi-formal configuration of microfinance and the provision of a range of financial and non-financial services to members based on joint responsibility (Yadav & Sharma, 2015).

Regarding the components of agricultural credits, Yadav & Sharma (2015) highlight direct credit, which includes short-term loans, medium-term loans, and long-term loans for agriculture and connected activities with direct responsibility for repayment. According to Gowhar et al. (2013), short and medium-term loans are provided by cooperatives, commercial banks, and regional rural banks for agriculture and allied activities. Whereas, long-term loans for agriculture are provided by rural development banks and primary cooperatives. Short-term agriculture credit enables farmers to buy inputs such as fertilizers, seeds, power, irrigation and the cost of the hired labor (Osuntogun, 1980; Adebayo & G, 2008). Short-term credit is practically for 6 months. However, long-term credit is oriented toward large investments such as irrigation pumps, tractors, and agricultural machinery (Anwarul & Prerna, 2015). While indirect credit allows the farmer to benefit from subsidized inputs and warehouse facilities. In this case, farmers are under indirect repayment responsibility through fertilizers dealers and

6

input suppliers. Figure 2 summarizes the components covered under the scope of institutional credit.

Direct Credit

Indirect Credit

Subsidized Inputs

Warehouse facility

Setting up

agribusiness centres

t

Credi

Agricul

ture

Short term loans

Medium term loans

long term loans

Figure 2: Components of direct and indirect agriculture credit Source: Yadav & Sharma (2015)

2.1.2. Financial institutions (FIs) in the Agriculture Sector

Financial institutions (FIs) are organizations that engage in the business of facilitating financial and monetary transactions. There are different types of financial institutions in a developed economy. Financial institutions cover also commercial banks, insurance companies, and brokerages firms. Furthermore, financial institutions can differ by size, scope, and geography. They offer a wide range of products and services such as transactions, deposits, loans, investment, and currency exchange.

In the agricultural sector, the main source of loans for smallholders' farmers and agribusinesses are other agribusinesses in the VC. However, farmers and agribusinesses can benefit from credit offered by FI. These providers of rural and agricultural finance can be broadly Banks (commercial, agricultural banks, state development banks), non-bank financial institutions (NBFIs, commercial MFIs, other non-bank lenders, and leasing companies), Not-for-profit MFIs, which tend to work with poorer clients, and Credit unions and agricultural cooperatives.

KIT and IIRR (2010) have shown that IFs have the capacity to develop new markets for all actors in the chain and make them bank customers. In the Global South, agriculture is the backbone of the economy, then the capacity to benefit the sector can significantly increase the actions of FIs.

2.1.3. Microfinance

Generally, the term MFI is given to non-profit organizations that depend on donations and grants to enable them to fulfill their primary social role of poverty alleviation (D'Espallier, 2012). Thus, in its limited concept, microfinance is the provision of micro-credit for small entrepreneurs who lack access to the formal financial system. Over time, Microfinance has developed from providing micro-loans for low-income people to collecting savings, micro-insurance, micro leasing, assisting with money transfer in relatively small transactions for disadvantaged people, and finally marketing and distributing client's products (Thai-Ha, 2020; Ferdinand & Asmah, 2012). MFIs play a key role to boost the social capital and the inclusion of disadvantaged populations and serve hundreds of millions of low-income borrowers (Morduch, 1999) because it allows access to credit to non-bankable people with all the advantages from banking services (Yunus & Jolis, 1999). The aim is to reduce poverty and develop economies, especially in Asia and Africa. Moreover, microfinance has enabled smallholders to improve the living standards of people, increase income and generate employment (Thai-Ha, 2020). A web-based report on microfinance shows a practical framework for understanding how MFIs function after a typical microfinance intervention using the illustration in Figure¹ 3. The main objective of microfinance is economic empowerment through the use of micro-credit as an entry point for overall empowerment.

7

Donors and Banks

	MFIs	State Banks

Micro Enterprise

Production Needs

Farm related

Micro Enterprise

Production Needs

Non-Farm related

Individual

Promotional work,
Formation
Implementing Org.

Credit Delivery,
Recovery,
Monitoring

Income generation

Economic
Empowerment

Individual

Figure 3: Development Process through Micro-finance

1 Source: (Ferdinand & Asmah, 2012, p. 76), Accessed: 18 July 2021

See https://journals.ug.edu.gh/index.php/gssj/article/download/1174/774#page=78

8

Source: Ferdinand & Asmah (2012)

Concerning the differences between these stakeholders, microfinance is a financial service that caters to the needs of the poor and micro enterprises, and it is typically a collateral-free short-term loan. In contrast, commercial banks generally do business with corporate clients, SMEs and individuals with higher incomes, and offer financing mainly based on collateral and repayment capacity. Alternatives include central banks serving the banking system. They facilitate cross-border transfers of money between banks and government institutions, both domestically and abroad. This flow is required while MFIs are complementary, not substitutes for, banks, donors, and state banks (Miguel & Silvana, 2007). In the figure above, we can see how effective collaboration between social welfare programs, MFIs, state Banks, and commercial banks may lead to greater poverty alleviation.

According to Ferdinand & Asmah (2012), this reduced process depends on microfinance's use to create a sustainable environment and more opportunities. Success in implementing microfinance is mainly linked to the ability of MFIs to meet the objectives of Donors and Banks in facilitating credit approval and the increase in the percentage of borrowers' repayments.

MFIs are always adopting different innovations to expand the delivery of microfinance in rural areas, Meyer (2007) announced several new products, technologies, and institutional connections intending to target rural areas and make financial services available for rural households. Some moneylenders use standing crops as collateral for loans. Others take assets as collateral for short-term farm and non-farm loans (World Bank, 2007).

Gonzalez & Rosenberg (2006) compiled a database that includes 2600 MFIs with 94 million borrowers. This database encompasses several micro-credit providers that are granted by a variety of FIs. The following table² 1 shows the approximate share of each type of FIs in the approval of micro-loans.

Table 1: Distribution of MFIs by institutional type

Institutional type Percentage %

State owned banks	30%
State owned Institutions	30%
NGOs	25%
Private banks and finance Companies	15%

2 Source: Gonzalez & Rosenberg (2006, p.2) Accessed 17 July 2021

See https://www.researchgate.net/publication/228276752 The State of Microfinance -

Outreach Profitability and Poverty Findings from a Database of 2300 Microfinance Institutions

9

Source: Gonzalez & Rosenberg (2006)

In the agricultural sector, various authors have concluded that microfinance boosts short-term agriculture investment, earnings, and consumption (Kaboski & Townsend, 2012; Mosley & Hulme, 1998). Unfortunately, while microfinance shows positive impact on Agriculture and Transformation, it has not solved the challenges r thatsmall-scale farmers face. Moreover, it has not shown a significant impact in terms of employment, increased income flows, and physical asset accumulation (Ferdinand & Asmah, 2012; Buckley, 1997; Coleman, 1999). In this context, some Scholars have enumerated several faults such as the high-interest rate, aggressive collection method, and driving people into debt (Thai-Ha, 2020). Besides, Diagne und Zeller (2001) have found following their study in Malawi that microfinance does not show any significant effect on household's revenue (Diagne & Zeller, 2001).

Hence, microfinance has its limitations and faces several challenges in lending to small-scale farmers. For this reason, continuous innovations are needed to enable MFIs to be more efficient in improving farmers' income, poverty alleviation, and economic empowerment while making service providers more sustainable over time.

The following section focuses on the Agriculture Value Chain Finance (AVCF) approach, which can complement and go beyond microfinance. The important difference here is that AVCF is linked to all actors and relationships in the chain in addition to transactions. Therefore, this concept can join several actors in microfinance. In addition, microfinance can be part of AVCF but must be with other financial services to address the different needs of the chain (KIT and IIRR, 2010).

2.2. AVCF definition

There is still no unified definition of AVCF. Different authors show an understanding with various characteristics on this subject. However, one of the most widely accepted definitions is the one formulated by Miller and Jones (2010), who define value chain finance as:» the flows of funds to and among the various links within a value chain" and distinguish between internal and external value chain finance. Likewise, authors from FAO and AFRACA (2020) defined AVCF as two internal flows of financing between chain actors directly within the VC and for financial service providers who use AVCF to lend money or to invest in one or more of the chain actors. However, the authors of KIT and IIRR (2010) have defined the VCF triangle, in which FI engages with the actors of the chain. This triangle is among FI, the seller and the buyer. The figure below illustrates the VCF.

Seller

financial institution

VCF
triangle

Buyer

10

Figure 4: Overview of value chain finance Triangle

Source: KIT and IIRR (2010)

This figure shows the payment, loan, and information and services flows between the financial institutions and the seller. Additionally, the payment and information flow between buyer and FI. Eventually, the flows of information and services and product between the buyer and the seller.

Complementarily, a study by Carroll et al. (2012) provides a pragmatic definition of AVCF:

...in the case of agriculture, the value chain may include (but is not limited to) input provision, production, processing, transport, storage marketing, and export.

Additionally, the Asian Development Bank (ADB) (2012) makes the definition of AVCF simpler:

...organized linkages between groups of producers, traders, processors, and service providers (including nongovernment organizations) that join together to improve productivity and the value-added from their activities.

Similarly, Zander (2016) presented the following definition»:

Value chain finance (VCF) denotes all financing arrangements within a specific value chain or from outside the chain. As the concept of value chains and their financing is broad and multifaceted, the terms `value chain' and `VCF' necessarily refer to a broad range of different instruments and mechanisms.

Finally, a recent study by Villalba, Venus, & Sauer (2021) explains Agricultural Value Chain Finance (AVCF) as:

A variety of products and approaches that allow stakeholders from a value chain to leverage social capital and satisfy the funding needs of the weakest actors. Cooperating within a value chain reduces risk, which can facilitate the

11

acquisition of financing from financial institutions, and other lenders at a lower

cost.

While no common definition has been proposed in the literature, ADB (2012) and Carroll et al. (2012) have the constituting element «provision», «processing», and «productivity» in common. They show that the chain starts from the raw material stage to the final consumer. However, other researchers explain this term as a variety of different products, mechanisms, and instruments used by different actors in the chain to initiate financing arrangements (Villalba, Venus, & Sauer, 2021; Zander R. , 2016).

There is agreement between the literatures when it comes to the flows of funds. (Miller & Jones, 2010; KIT and IIRR, 2010; AFRACA and FAO, 2020). In addition, there are definitions of AVCF for which the authors partially agree such as internal and external value chain (Miller & Jones, 2010; AFRACA and FAO, 2020).

The authors agreed that Internal Value chain finance takes place within the value chain such as when an input supplier provides credit to a farmer, or when a lead firm advances funds to a market intermediary. External value chain finance is that which is made possible by value chain relationships and mechanisms: for example, a bank issues a loan to farmers based on a contract with a trusted buyer or a warehouse receipt from a recognized storage facility.

Other authors defined AVCF as a triangle, in which, an agreement between the actors (FI, seller and buyer) is made around the product, the need for financing, the sharing of information, the method of communication, and finally the way of risk management (AFRACA and FAO, 2020). This agreement according to KIT and IIRR (2010) allows the development of the value chain in three different ways:

a) Ensuring liquidity for the actors of the chain

b) Creation of new chains

c) Investments in existing chains

This highlights how general financing of agriculture works, (new investments, reinvestments, and financing of current assets) and is a useful typology for value chain development.

2.3. AVCF Challenges

About the challenges of AVCF, the authors have listed several constraints that buyers and suppliers face in lending to farmers as shown in the following Table 2.

12

Table 2: The challenges of the agriculture finance according to the literature review

Challenges	Author(s)
	Pearce (2003)	Langen bucher (2005)	Honohan and Beck (2007)	Meyer (2011)	IFC (2012)	AfDB (2013)	Klonner and Rai (2015)	Herliana (2018)	ISF (2019)
Financial exclusion of farmers		X	X	X	X	X	X	X	X
information asymmetries	X	X		X	X	X	X	X
Transaction cost	X	X			X	X	X	X
High fees					X	X	X	X
insufficient amounts of credit					X	X			X
Low infrastructure, distant location		X			X
Low Education of farmers								X
Fluctuating Production								X
Lack collateral								X
Inefficient Market					X

The challenges cited in most literature studies are the following:

Financial exclusion of farmers: according to Langenbucher (2005) the causes of financial exclusion for the small farmer from the supply side are the lack of robust business models, and limited access to equity capital. Likewise, the key features that influence value chain finance (VCF) are the high incidence of informality (lack of documentation and contract), the intermediation deficiency (high-interest rate and minimum deposit), and the dominance of the banking sector (lack of information about credit worthiness of potential clients, weaknesses of the legal system, and high degree of corruption and inefficient bureaucracies. (Langenbucher, 2005; Honohan & Beck, 2007; Meyer R. , 2011). Meanwhile, formal institutions are less interested in financing the agricultural sector due to several constraints and subsequently obtaining formal credit is a complex procedure (Herliana, Acip, Qorri, Qonita, & Nur, 2018).

Information asymmetries: in accordance with Pearce (2003) and Langenbucher (2005), which makes the agricultural environment very complicated, are a lack of information about

13

14

smallholders' farmers and risk of agricultural activities. In addition, this sector is complicated due to the unique problem of the agricultural sector, the inadequate policy, regulatory environments, and asymmetric information (IFC and World Bank, 2012; Herliana, Acip, Qorri, Qonita, & Nur, 2018). In addition, several studies cited that FIs are not aware about farmer's credit worthiness, and therefore, smallholders' farmers are left when FIs try to mitigate risk (Klonner & Rai, 2005; IFC and World Bank, 2012; Pearce, 2003). Thus, banks were not investing adequately to understand the demands and nuances of value chains (VCs). This lack of information leads to the design of financial products that are not suitable for rural activities (AfDB, 2013).

Transaction cost: can be viewed from a financial point of view as a difference between the price a broker pays for a security and the price the buyer pays (Cheung, 1987). Most authors agree that transaction costs are one of the most important constraints that FIs have to deal with (Langenbucher, 2005; IFC and World Bank, 2012; AfDB, 2013). According to Pearce (2003), buyer and suppliers face difficulties in lending to farmers. Among these difficulties, he mentioned transactions cost as a major problem.

Following the bibliographical study concerning the constraints, the authors show a partial compromise about the following challenges:

High fees: as defined in the report of the AfDB (2013), small farmers who already have access to loans find the terms very rigid and the fees too high, which causes an increase in production costs. As a result, they borrow money from family and friends, or money lenders who charge high interest and limit their potential to expand (IFC and World Bank, 2012; Herliana, Acip, Qorri, Qonita, & Nur, 2018).

Insufficient amounts: according to ISF Advisors (2019), small farmers in South Asia, sub-Saharan Africa and Latin America need about USD 240 billion for agricultural and nonagricultural finance. Service providers are not able to meet this demand and the latest data collection indicated that financial service providers provided only USD 70 billion distributed as follows: USD 30 billion by value chain actors, USD 21 billion by formal financial institutions, and USD 17 billion by informal and community-based financial institutions. This shows that 70% of the global demand of small farmers remains unsatisfied, the equivalent of USD 170 billion and the contribution of FIs remains minimal (ISF Advisors and Mastercard Foundation, 2019). Additionally, most loans are short-term which cannot improve the output and income (AfDB, 2013). Moreover, lenders confront irregular payments and slow rotation of capital (IFC and World Bank, 2012).

Low infrastructure and distant location: While Langenbucher (2005) in this context has listed the lack of appropriate risk mitigation and infrastructure, and no branches or limited

network in rural areas, the authors of IFC (2012) enumerated the low population densities, low infrastructure, distant locations, and the inefficient market, which can only worsen the situation and decrease profitability.

Finally, other challenges were listed separately and show less compromise between the others:

Low education of farmers one of the barriers to agricultural finance is the low education level of small farmers (Herliana, Acip, Qorri, Qonita, & Nur, 2018). This low literacy level of cultivators is the main cause of the limited access to information which translates to low efficiency in resource management and then low productivity crops. This low productivity is due to the lack of knowledge about the right proportion of Inputs (seeds, chemical fertilizers, and pesticides). Besides, in a present changing environment, farmers are unable to compete which traps them in a vicious circle of poverty.

Fluctuating Production Most banks avoid financing agriculture on the ground of fluctuating production and uncontrolled price risk (Herliana, Sutardi, Aina, Himmatul, & Lawiyah, 2018). The main causes of the fluctuating production are climatic factors such as: water, light, rainfall, and temperature which lead to unpredictable food production (Gilbert, 2010). In addition, ecological land change and cultivated land intensity will make the production more complicated (Xie, He, & Xie, 2017; Xie, He, Zou, & Wu, 2016).

Lack of collateral Herliana (2018) showed that FIs avoid financing small farmers following the low profits and lack of collateral. Besides, rural producers do not generally have assets that IFs are willing to accept as collateral. In addition, pledges of agricultural assets have often been insufficient to ensure credit recovery, thus threatening the sustainability of development Financial Institutions (FAO and ALIDE, 1996).

Inefficient Market According to IFC authors (2012), unstable market prices aggravate the financial situation of small farmers. Food demand is constantly rising due to growing population. however, food supply is dropping due to increasing production cost. This imbalance in food demand and supply is the primary cause of the price volatility (Lu, 1999). Besides, continuous changes in input and output prices reduce the income of small farmers (Ijioma & Osondu, 2015).

2.4. Competitiveness of agricultural finance

According to Abid et al. (2020), to achieve the objective of reducing transaction costs (TC) and risk management, several factors should be considered by organizing chain actors in VCF. Among these factors³, we can mention; Agility, Innovativeness, Information sharing, Trust,

3 Source: Abid et al. (2020)

15

Information & Communication, and Contractual Governance (Abid, Jie, Aslam, Batool, & Lili, 2020).

Information Sharing

Agility

Trust

Competitiveness

Information & Communication

Innovativeness

Contractual Governance

Figure 5: Considered factors to reduce TC and risk management in agricultural finance

i. Agility: which is the ability to respond to changing customer needs and satisfy
the unique needs of each customer with flexibility and adapt to unforeseen circumstances (Abid, Jie, Aslam, Batool, & Lili, 2020). Then, agility in the value chain allows the FIs to respond quickly to the needs of customers and then ensure competitiveness in the chain. This accelerates the financial flow in the VC and reduces the intervention of informal money lenders (Ellis, 1992). In the same scope of work, Meyer (2007) emphasized the important role of the intervention of FIs in identifying the unmet demand and to find where the lending cost and risks are lower. He mentioned that this follows up requires more analysis to design appropriate product, improve lending capacity, and make diversified loans portfolio. Therefore, Agility in the VCF improve competitiveness with respect to financial product and innovation (Swafford, Ghosh, & Murthy, 2006).

ii. Innovativeness: is a system in which VC actors provide new products and
services to increase customer satisfaction (Simon & Yaya, 2012). The authors of IFC (2011) and Hult (2004) found that FIs need to develop new credit skills and policies, credit scoring and rating tools, as well as portfolio monitoring practices to provide high value customers. The way of lending credit to farmers is parametric, while it is based only on a few parameters due to the lack of collateral. This should be improved, developed, and replicated. Generally, FIs

16

with a high degree of innovation can adapt to every change in the environment and ensure more competitiveness within the VC. Moreover, according to Kaufmann (2000), FIs can improve their ability by strengthening the process of product innovation by organizing producers and buyer's relationship. Equally, According to Röttger (2015), FIs should understand the risks and opportunities of the sector and link loan with insurance products.

iii. Information sharing: according to Brown (2010) and Pagano (1993),
Information sharing describes how the actors in the chain react to each other over time. Information sharing is very important as it improves market competitiveness, credit allocation efficiency, reduces asymmetric information, and increases lending volume. Furthermore, the sharing of information helps FIs to address asymmetric information using the network, incorporate the collaboration with VC actors to design innovative product (Miller C. , 2012; Röttger, 2015), in this regard, the authors of KIT and IIRR (2010) and Kim (2017) highlighted that information sharing return the relationships in the chain smooth and simplified and build better partnership. Moreover, the cost of credit screening, monitoring and enforcement are reduced due to chain actors, which take part in this procedure and do this work themselves and make agile the VCF (KIT and IIRR, 2010; Kim & Chai, 2017).

iv. Trust: Trust is about how reliable and credible the partners in the value chain
consider themselves to be (Abid, Jie, Aslam, Batool, & Lili, 2020). Trust is therefore considered as an important element to overcome unexpected situations and to act in situations perceived as risky (Song, 2018). According to Jones et al. (2015) trust between stakeholders is an essential requirement for successful management of financial flows in the value chain. In this context, Miller, and Jones (2010) mention that trust between producers and buyers is one of the key factors to mitigate risks. Thus, the adoption of an agile approach VCF is based on a better level of trust (Svensson, 2001).

v. Information & Communication: according to Ali et al. (2011) and Imtiaz et al.
(2015), Information and communication enhance the creditworthiness of smallholder's farmers, reinforce partnerships between value chain actors, minimize the cost of stakeholder interaction and the risk related to the VCF. For more effectiveness and risk reduction through information and communication, the authors of IFC (2012) have mentioned the development of insurance products, such as crop and weather insurance products, credit life product, and emerging health and FIs should rely on cash flow, saving or group guarantee, then within their value chain and take advantage of all information to develop

17

adequate product that combine financial and non-financial services. Besides, FIs should improve their capacities to assess farmer credit, develop insurance and risk-sharing, and identify opportunities to increase their level of comfort and reach more farmers (IFC and World Bank, 2012). Likewise, Röttger (2015) highlighted the importance of relying on the recommendations of group leaders and extension agents in the aim of linking loan with insurance products, increasing loan amounts, and declining interest rate. Zhao et al. (2019) have found that the use of information facilitates innovation in financial services, lower costs, monitor risks, and allow for agile financial flows.

vi. Contractual Governance: is the degree to which a contractual partnership is
legislated by a formal contract specifying formal rules, responsibilities, and duties (Zhou & Poppo, 2010; Cao & Lumineau, 2015). Moreover, it supports innovation-based coordination and strict collaboration between actors. This cooperation between value chain partners, according to Anna Grandori (2019), regulates transactions and the pooling of resources, and thus, procedures for innovation. According to the authors of the AFRACA (2020), AVCF offers several advantages for chain actors. Starting with producers, AVCF makes it easier for them to get credit because of the lack of collateral. Then, for agribusinesses, AVCF strengths the buying-selling relationship and allows market expansion. Finally, for FIs and investors, AVCF reduces transaction costs, improve repayments, and mitigates risk due to more suitable financial products. Due to the absence of FIs in most rural areas because of cost and risk of agriculture lending are too high, the authors found that FIs can take advantage of the collaboration in the VC and work with companies rather than directly with farmers to deal with these risks... Eventually, the public sector has a crucial role for a suitable policy and banking regulations that allow the application of new approaches, and infrastructure.

2.5. Determinants of agricultural credit

Several studies have been done in the past on the identification of the determinants of agricultural credit and the factors that significantly influence the decision (Akpan, et al., 2013; Salami & Arawomo, 2013; Yuan, Hu, & Gao, 2011), many variables (factors) have been identified in the literature leading researchers to analyze their impact on the decision of credit accreditation (Abid, Jie, Aslam, Batool, & Lili, 2020).

To start with Meyer (2007), FIs need to ask farmers before lending, such as others engaged in economic activities for the households, cash inflows and outflows, source of income for

repayment, and structure of the given loan. This can help FIs to determine the amount to be lent to the specific VC, and the method to mitigate risks. Thus, this evaluation allows determining the creditworthiness of farmers, the types, terms, and conditions of financial products needed to meet these demands. Other studies have also identified specific factors for the allocation of credit such as education, marital status of the household, contact with extension agents, years of experience in farming, land size, gender, etc. (Aliero & Ibrahim, 2011; Dzadze, Osei, Aidoo, & Nurah, 2012; Akpan, et al., 2013). Among other socio-economic factors, being a member of a cooperative plays a key role in access to credit (Ijioma & Osondu, 2015). A recent paper in Pakistan has shown also that health status remains one of the determinants for credit accreditation (Saqib, Kuwornu, K.M., Panezia, & Ubaid, 2018).

According to Gammage (2013), access to bank finance is determined by a number of factors such as ownership type, age of the firm, sector, and location of the business, assess tangibility, firm performance, availability of audited financial statements, gender of the owner-manager and perception of the owner-manager of access to bank finance, and characteristics of the borrower at the time of evaluating loan applications. According to Abid (2020), the most common factors are literacy, size of land, marital status, and distance to a lending institution, age of the borrower, caste, religion, and value of assets held by the household. Another study in Nigeria found a significant relationship between gender, marital status, the lack of a guarantor and access to credit (Ololade & Olagunju, 2013). In addition to gender, agricultural experience, education level, farm size, and income, household size and availability of collateral have a meaningful effect on loan accreditation for farmers (Abbas, Yuansheng, Feng, & Liu, 2017).

From the review, these factors were classified into three groups based on common characteristics, characteristic linked to the farmer (Gamage, 2013; Ololade & Olagunju, 2013; Ijioma & Osondu, 2015), others linked to the farm (Aliero & Ibrahim, 2011; Gamage, 2013), and finally those linked to the economic activity (Meyer R. L., 2007; Gamage, 2013), as shown in the Table⁴ 3.

18

4 Source: The Author

Summary of different research on the determinants of access to credit

Table 3: Determinants of credit Access

19

Farmer		Farm		Economic Activity
·	Education	·	Land Size	·	Cash inflows and
·	Marital status	·	Ownership type		outflows
·	Household size	·	Age of the firm	·	Source of income for
·	Years of experience	·	Sector		repayment
·	Being member of a cooperative	·	Location of the business	·	Availability of audited
·	Health status	·	Firm Performance		financial statements
·	Gender	·	Distance to a lending	·	Value of assets
·	Characteristics of the borrower		Institution	·	Availability of collateral
	(caste)
·	Religion
·	Guarantor

2.6. Literature on AVCF

2.6.1. Gap of the Literature on AVCF

Recent studies have focused on the determinants of the sources and amount of agricultural credit (Yadav & Sharma, 2015), and on evaluating the access to finance at the farmers' level (Gamage, 2013). However, few researchers have taken into consideration the contribution of financial institutions (FIs) and the identification of the supply-side determinants of agricultural credit (Meyer R. L., 2002; Yadav & Sharma, 2015). This could be an area of further potential investigation, mainly, in new approaches such as AVCF (Villalba, Venus, & Sauer, 2021).

According to Zander (2016), the research available on AVCF is relatively scarce due to a number of major concern that restricts the knowledge about non-bank based agriculture value chain financing. First, the hesitation of the private sector and agri-businesses to share their information with others due to confidentiality issues. Second, the hesitation of FIs to share operational and performance details of their agricultural portfolio due to confidentiality. Finally, the hesitation of analysts and authors to disclose details on outcomes and impacts, since the initiatives are new and the results are not sufficiently robust.

20

With respect to the adequate information on this topic, there is a significant gap in the available literature overall. The influences of the AVCF are little reported in the literature, even though external facilitation needs strict monitoring of the impact of the AVCF, especially for small-scale producers

2.6.2. Available literature on AVCF

Regarding the Types of the existing literature on AVCF, there are three types of documents available (Zander R. , 2016):

a) Normative information and guidance

b) Descriptive documents with facts and figures

c) Anthropological and sociological studies

Concerning the studies available on the development of AVCF in the literature, Miller and Jones (2010) discuss 5 cases of AVCF involvement; 1. Numeric project in Kenya and Tanzania 2. Inventory credit system Niger 3. Integrated agri-business finance model 4. Technological innovations in Kenya 5. Integrated Agro-food in India. The analysis of actual cases and results in this paper does not show the aspect of the implications of AVCF on agricultural or financial sector development (Zander R. , 2016).

KIT and IIRR (2010) address the financing gaps that exist within agricultural value chains. This book presents detailed case studies. These include sections on results, impact, threats and challenges, and lessons learned. Among the cases discussed we can mention; Credivida in Peru, BASIX group in India, K-Rep Group in Kenya, Organizations supporting the soybean value chain in Ethiopia, UCPCO and Fondo de Desarrollo Local in Nicaragua, and Pro-rural In Bolivia.

A publication of Inter-American Development Bank (2010) has highlighted the relationship between AVCF and local financial and agricultural sectors. This report compared two value chains in Nicaragua (dairy and plantains) and Honduras (plantains and horticulture) and included relative observations on mechanisms used within the value chain financing (Coon, Campion, & Wenner, 2010).

Another FAO publication of Da Silva and Rankin entitled «Contract Farming for Inclusive Market Access» presented case studies of different value chains of cacao, sugar, oil palms, and other plantations crops which work with international markets under contract farming between lead firms and cooperatives. This report emphasized the importance of this instrument which is a very interesting development tool with growing expectations for VC promoters (Da Silva & Rankin, 2013).

21

A recently published discussion paper from the Deutsches Institut für Entwicklungspolitik, discusses the implications of the AVCF arrangements for agricultural sector development and presents strategies that reduce the risks for financial sector growth, with the aim of better reaching priority segments of the rural population such as small-scale farmers and rural micro and small enterprises (Zander R. , 2016). This paper illustrates in detail the following 4 case studies: 1. accelerating production and post-harvest infrastructure in Rwanda 2. Fostering AVCF in Ethiopia 3. COMPACI project in Zambia 4. KELIKO farmer association in South Sudan.

The following table shows the different publications available from which we have selected all the FIs that have experience in the application of the AVCF. These FIs will be contacted for the survey later

Table 4: Number of FIs mentioned in the literature review

Publication	Authors	Year	No. FIs
AVCF: Tools and Lessons	Miller and Jones	2010	45
Value Chain Finance	KIT and IIRR	2010	28+(2^5)
Financing Agriculture Value Chains in Central America	Coon, Campion and Wenner	2010	8+(2^7)
Agricultural finance for smallholder farmer	Daniela Röttger^8	2015	8
Contract Farming for Inclusive Market Access	Da Silva and Rankin	2013	3+(1^7)
Risks and Opportunities of Non-Bank- Based Financing	Zander Rauno	2016	17
Consultative Group to Assist the Poor^9	N/A^8	2015-2016	44
World Council of Credit Unions^9	Elissa McCarter^8	2021	48
Total of FIs mentioned in the literature review:			201

5 Financial institutions mentioned by Miller and Jones (2010)

8 FIs cited in bibliographic studies where the term AVCF is not mentioned

⁹ Source: CGAP, 2016 Assessed: 31 July 2021 See: https://www.cgap.org/small holders data portal/ ⁹ Source: WOCCU, 2021 Assessed : 25 August 2021 See: https://www.woccu.org/

22

3. Methodology and Data

Our research aims to examine how FIs can offer finance to support the actors along the value chain utilizing an AVCF approach. To our knowledge, no previous research has investigated how FIs use AVCF to support their clients.

3.1. Description of the study

The study presented in this master's thesis is based on data collected by the author between July and September 2021 from financial institutions' websites in several developing countries in Africa, Asia, Oceania, and South America. It aims to highlight the extent to which the financial institutions can attract smallholders' farmers and agribusiness to benefit from the study presented in this master's thesis heir services.

In total, 106 countries are included in this database [Annex B]. Among them are 98 developing countries and 8 developed countries, respectively Belgium, France, Germany, Italy, Luxembourg, the Netherlands, the UK, the USA.

Figure 6: Geographical location of the financial institutions

This database contains not only general information about the FIs, but also specific information detailed in the following section. This database will be used in the future to conduct a survey to enquire about the application of AVCF by FIs in several languages for the convenience of the participants.

23

3.2. Building the database

Increasing numbers of FIs in the global south are involved in the agricultural sector and report their progress in recent years (Gonzalez & Rosenberg, 2006). The following database encompasses initially agricultural Banks. Afterward, it has been extended to reach Commercial Banks, state development banks, non-bank financial institutions (NBFIs), leasing companies, microfinance institutions (MFIs), credit unions, insurance companies, NGOs, international financial institutions, credit unions, and agricultural cooperatives. The purpose of investigating different FIs with different branches is to enquire if they apply agriculture value chain finance.

The research purpose consists in reaching the maximum number of FIs in the Global South and collecting the maximum amount of information about these FIs to facilitate sampling later. Qualitative methods were employed to identify FIs, their type, foundation year, number of branches, and if available agricultural credit, gender programs, and digital solutions. I used several strategies to create the database starting with the common databases available, websites, and especially the FIs cited in the bibliographic studies. The database sources are:

I. Finance in Commons⁶: a website that provides limited information on many
Commercial, Housing, agricultural, Export, and Local Banks. This database contains 39 agricultural banks worldwide, with $ 1,221,281 MIs USD total assets and $ 49,306 MIs USD total equity. The database is collected as part of the research program «Realizing the Potential of Public Development Banks and Development Finance Institutions for Achieving Sustainable Development Goals» launched by the AFD, INSE, and IDFC. The database is published annually and can be found online at PDBs database The rest of the information includes net Interest Income, net Profit, and level of income (AFD, INSE & IDFC, 2021).

II. Case studies from the Literature Review: All FIs mentioned in the literature
offering finance with a VC approach were included (Table 4; section 2.8.2). These FIs show at least one application of the AVCF concept in one or more countries (case of an international institution). Moreover, other institutions cited in surveys with MFIs, studies on the agricultural credit, research on access to bank finance, and paper on credit policy were taken into consideration for the database.

6 Source: AFD, 2021 Accessed: 20 July 2021 See https://financeincommon.org/pdb-database

III. FIs' websites: For more information on the FIs, their official website is
consulted to determine the contact (Email), the responsible person or the CEO, and the availability of agricultural loans or agribusinesses loans.

IV. Internet search engines: to identify all FIs that fund agricultural projects in
the French and Arabic speaking countries we have searched for each country with these terms «agricultural», «banks», «MFIs», «financing», «agriculture», «agricultural», «sector», and «country name».

The following table shows the sources of FIs for the composition of the database:

Table 5: Composition of database

FIs Sources	No.
Finance in common	37
Case studies from the Literature Review	201
FIs' websites & Internet search engines	109
Total FIs	347

I

II

III & IV

24

After eliminating double counting, these sources have been consolidated into a single database that includes about 347 FIs. The database covers the country of origin of every FI, acronym, foundation year, the number of branches, name of CEO if available, E-mail, and website. These contacts were determined with the aim to survey the role of financial institutions in Value Chain Finance in the Global South.

In addition to the FIs, the database covers development funds, NGOs, and international financial institutions such as IFC and IFAD which finances agriculture and supports the value chain approach in many countries. There are thousands of other financial institutions that have been financing the agricultural sector for hundreds of years without being considered participatory in the AVCF (Zander R. , 2016). Furthermore, the database captures only a small fraction of funding bodies and relatively little information about the participating institutions.

25

3.3. Statistical Analyses

3.3.1. Qualitative Analyses:

The term qualitative data refers to any data which is not numerical. Our case involves geographical variables, the types of institutions, the availability of agriculture loans, gender, and digital solutions. Depending on our objectives and research questions, we will focus only on thematic analysis from the six most popular qualitative methods listed below.

Table 6: Most popular Qualitative Analysis method

Qualitative Analysis Method

1 Content Analysis

2 Narrative Analysis

3 Discourse Analysis

4 Thematic Analysis

5 Grounded Theory

6 IPA: Interpretative Phenomenological Analysis

Thematic Analysis: This method focuses on themes and patterns and examines meanings in a dataset. Essentially, this kind of analysis draws similarities from a large body of data that is often quite large and arranges them into themes. By focusing on specific themes, the study makes sense and has more meaning. It also allows us to compare several sets of variables. In this case, it would be feasible to review a few hundred FIs' websites to identify which one deals in agricultural credit. This would be an opportunity to find out also about the offers, services, and orientation of FIs.

3.3.2. Quantitative Analyses

In quantitative analyses, we analyze data that is numerically based, or data that can be translated literally into numbers without losing the context, such as the gender in our case. This method is employed in the current thesis for three purposes; namely, to measure differences between groups, to analyze relationships between variables, and to test hypothesis. These tasks will require a combination of descriptive statistics and inferential statistics. The differences between the two are that descriptive statistics focus on describing the contents of the sample, while inferential statistics seek to make predictions about the population.

The population in this context refers to the full set of financial institutions that we are interested in researching. However, we can only have access to a subset of these institutions. The reduced sample size is explained in detail in Section 3.5.2 of the report. In order to answer our

26

research question, we will use both types of quantitative analysis. The following table explores both types of statistics.

Table 7: Quantitative statistic types

descriptive statistics inferential statistics

1) Mean (average) 2) Median (midpoint) 3) Mode (most common) 4) Standard deviation 5) Variance 6) Skewness

1) Correlation analysis

There are many advantages to using descriptive statistics. Among them are that they provide a macro and micro picture of data, facilitate spotting errors and anomalies, and suggest which inferential statistics to use. In the other hand, inferential statistics determine predictions between two or more groups and most important relationships between two or more variables.

3.3.3. Cluster Analysis

Using Excel functions will enable us to do descriptive and inferential statistics, but we will need to use the software RStudio to run cluster analysis and process complex functions.

A cluster analysis or clustering is a data mining process in which a set of observations is grouped so that elements belonging to the same group are more similar than those in other groups. These groups are known as clusters (Vandeginste, et al., 1998). These are the two methods used to analyze clusters: K-means Clustering and Hierarchical Clustering. Here's how both methods work:

a) K-means Clustering Method: This algorithm is usually known as K-means, which is based on centroid values. The idea is to define k centroids, one for each cluster. After this, each point is associated with the nearest centroid within a data set. In the absence of pending points, a premature grouping is done in the first step. Here, the algorithm relates the new centroids to the new cluster centers determined from the previous step. The algorithm stops when no further changes are observed between one iteration and the next (Vandeginste, et al., 1998).

b) Hierarchical Clustering Method: It is expressed as a tree, also known as a dendrogram, and it involves the development of a hierarchical classification. It involves the division of large groups into smaller ones. They are then split again until there is only a single meteorite in each group (Vandeginste, et al., 1998).

3.3.4. Data Types and Variables

Along with the bank's name, the continent, and the acronym, there are ten additional variables highlighted in the search. Of them, seven are completely accessible on the official websites of the financial institutions, and three are partially accessible. The table below shows the variables and the method for finding them.

Table 8: Data types

Variables Search Method Name/ Scale

Continent Institutional Type Foundation year Number of Branches Agricultural Loans Gender

Digital solutions

Total assets

Total Equity

Gross Loan Portfolio

checking the institute's website C 0-3

checking the rubric «about us» T 0-11

checking the rubric «our history» F Year

checking the rubric «locate us» B Number

Checking the rubric «Products and Services» AL1, AL2 0-1

Checking the rubric «our value» and «events» G1, G2, G3 0-1

Verifying if «online banking» or other e-products DS1, DS2, 0-1

available DS3

investigating the most recent «annual report» AS $

investigating the most recent «annual report» EQ $

investigating the most recent «annual report» LP $

To capture the variety of information, several attributes (dummy variables) and sub-variables have been assigned to the seven previous variables according to the following tables.

Table 9: Continent Attributes

Continent Attribute

Africa	0
Asia	1
South America	2
Global	3

Each continent (C) has been assigned a unique attribute between 0 and 3 as show in the previous table.

Table 10: Institutional Type Attributes

Institutional type Attribute

0

1

2

3

4

5

6

27

Agricultural Bank Commercial Bank cooperative

Credit union

Development Bank Insurance company

International FI

International Fund Islamic Bank

MFI NGO

7

8

9

10

state development bank 11

Each Institutional type (T) has been assigned a unique attribute between 0 and 11 as show in the previous table.

Regarding Foundation year (F) Number of Branches (B), As the data are numerical, they will be analyzed directly.

Table 11: Agricultural loans Attributes

Agricultural loans Attribute

Farmer credit	AL1
Agri-business credit	AL2

Farmer credit, AL1: The financial institutions included in this database provide a range of financial products, such as consumer credit, advances, salary advances, mortgages, car loans, or asset acquisitions as well as cashback loans. Apart from these loans, some FIs offer specific services and products to farmers, including Farm input loans to meet the need for financing inputs (such as seeds and chemicals), machinery, labor, and harvesting. Further, agricultural credit tailored to meet any economic need of the farmer, such as the purchase of merchandise or other income, with flexible amounts and long-term terms and fast processing is offered. During data processing, we will assign "1" to AL1 where institutions offer agricultural credit to small-scale farmers, and "0" to AL1 when no such credit is offered, or the offer is restricted to the above-mentioned non-agricultural credit.

Agribusiness Credit, AL2: This is a loan designed to help support the value chain players in the agricultural inputs and equipment manufacturing, stocking, importing, exporting, trading, and storing sectors. The credit facility is intended to facilitate agri-business enterprises and finance working capital and operational needs across the value chain. Beneficiaries may include agri-dealers, agro-processors, input manufacturers, and agro-importers and exporters. If there is an agribusiness credit, a "1" will be assigned and if there is none, a "0"

Table 12: Gender Attributes

Gender Attribute

28

Credit facility for women

	G1
Career development to female staff	G2
Gender Programmes	G3

Credit facility for women, G1: Many financial institutions have developed direct lending programs targeting women in the agricultural value chain. The objective is to finance women-led small and medium enterprises, strengthen their inclusion in agriculture, and encourage them to engage in commercial activities. In this case, a «1» will be assigned to G1, and «0» otherwise.

Career development, G2: Some financial institutions promote gender equality at work, which means that women can enjoy the same rewards, opportunities, and resources as men at a company, which include equal pay and benefits for comparable jobs with comparable responsibilities, equal opportunities for promotions and career progression, and equal consideration of needs. G2 will be assigned "1" in this instance and «0» otherwise.

Gender Programmes G3: Other banks offer gender programs and celebrate events such as International Women's Day, sign agreements with associations protecting women's rights by offering specific services. In addition, they allow women to participate in a draw to win a savings account or a scholarship dedicated to the education program abroad.

Other initiatives are available for women who excel in their studies to participate in workshops with senior executives and receive answers to their bugging questions. This aim is to empower youth females in taking a strong stance on their future. G3 will be assigned «1» in this case and «0» otherwise.

Table 13: Digital Solutions Attributes

Digital solution Attribute

Online Banking E-Products:

DS1

DS2

Online Loan Application

DS3

Online Banking DS1: Some financial institutions offer a secure online banking platform that could be used to access banking services efficiently. The portal offers clients the ability to review account balances, obtain statements, send money, pay bills, and review transaction history. DS1 would receive a "1" if this were available; otherwise, it would receive a "0".

29

E-Products: A variety of products are available to clients to keep them informed and simplify banking. Email banking, in which the account history over a specific period is sent. Mobile banking features include SMS alerts for withdrawals and deposits, as well as debit card transactions, funds transfers, and other transactions. Users can also access and manage their savings accounts and transfer money from their account to a beneficiary or anyone else using mobile banking. A "1" will be assigned to DS2 if one of these facilities is available, and an "0" if not.

30

31

Online Loan application: In several financial institutions, clients can apply for a personal loan online, securely, and easily. The process is fast and secure. Once the loan is submitted, the details of the facility fees and due date will be displayed, and SMS confirmation will be sent. DS3 will be assigned "1» in this case

Concerning international financial institutions, the number of agencies is being replaced by the number of countries in which it operates or the number of foreign offices. For Gender and Digital Solutions, it was checked whether the project focuses on women or presents a digital footprint of its approach. Lastly, the total assets and capital are substituted by the total amount granted by the institution for the project.

Concerning the last three variables (Total assets, Total Equity, Gross Loan Portfolio) and due to the lack of these information in several financial institutions, we will not consider them during the analysis.

3.3.5. Other components of the database

Additional data in the database include the institute's website, its general email address, the name of a CEO/manager who works in the institute, and if available, his personal contact information. The last cases are attributed to mentions of the institute in the literature and contact information of the author.

3.4. Structure of survey

Quantitative methods were used in a survey in order to measure the role of FIs in facilitating the AVCF in the Global South (GS). The purpose is to determine the financial products, credit screening and scoring for agriculture loans, mechanisms, and Instrument used by FIs in the aim of reducing TC, managing risks, and assisting small-scale farmers to access financing.

Online Survey: conducted in the Global South (GS) with FIs which apply and do not apply the AVCF. To define the survey questions, an extensive and careful review of the scientific literature will be carried out to check standard questions that have been used and employed by major survey, then develop the most appropriate questions. The period of the survey was in July-September 2021.

The survey contains 5 sections:

1) General information about FIs

2) Economic information

3) Agricultural credit: screening, scoring and monitoring

4) Agricultural finance within value chains

5) Financial product & Instrument employed

A standardized survey was designed using the platform Qualtrics used to create this survey, number questions, generating test responses and reviewing survey accessibility. Therefore, to facilitate data processing and for a quick tabulation of results, only multiple-choice questions will be used, and before starting the survey, a pretest of the survey and the pre-coding of these closed ended questions will be carried out. Finally, the data collected was analyzed using Microsoft Excel and R for result generation. This resultant analysis of the data will make it possible to draw conclusions about the effectiveness of the use of the Agricultural Value Chain Finance (AVCF) by financial institutions (FIs). This assessment provides evidence to decision-makers, development agencies, and all types of rural FIs and shows priority areas that require intervention regarding the accreditation of credit for small-scale farmers.

3.5. Sample Design

3.5.1. Sampling frame

This database includes several independent financial institutions from different provenances in the Global South with a total sample size of 347 institutions. It was designed to produce reliable estimation at the international level. Accordingly, the universe for the database includes FIs mentioned as per criteria identified in table 5; FIs from the platform Finance in Common, FIs from literature reviews on AVCF, FIs in other surveys and studies that do not include the AVCF methodology, and FIs from the internet search engines.

According to Dados & Connell (2012), The term "Global South" commonly refers to regions in Latin America, Asia, Africa, and Oceania. It is part of a family of terms, including "Third World" and "Peripheral," that refer to regions outside of Europe and North America that are considered mostly low-income and often politically or culturally marginalized. The use of the term «Global South» reflects a shift from a central focus on development or cultural difference to a focus on geopolitical power relations. Our sampling frame for the database included countries around the globe with at least one FI that funds the agricultural sector in the Global South even if the FI is not a member of this zone. The next figure shows the classification of the chosen countries by continents.

Database: 347 financial institutions across the 5 continents:

Africa

Sample FIs: 200

Asia

Sample FIs:

52

Latin
America

Sample FIs:

63

Oceani

a

Sample FIs:

16

Global

Sample FIs:

16

32

Figure 7: Geographical distribution of financial institutions

3.5.2. Sampling techniques

The total sample size of 347 FIs was divided into two categories based on their source. The categories are FIs encountered while conducting the bibliographic study and FIs obtained from other sources (Google & Platform: Finance in Common). It was decided to eliminate the second group (FIs not cited in the literature) as a means of reducing the sample size and facilitating processing. As a result, the sample (202 FIs) will be considered for time considerations. The exemption will apply only to financial institutions that lack an official or functional website. Due to the different orientations and types of agricultural interventions of these FIs, the differences in information available, the difficulty of gathering the same information under the same categories chosen, and above all to give meaning to the descriptive analysis, the final data sample will be reduced by 57 FIs.

In this case, the snowball effect will be used to construct this database. The database is then finalized by separating each FIs through the source, official website & Information available, and type of intervention in the agricultural sector. To account for the unavailability of the official website, missing information about the FIs, the target sample size was decreased to 144 FIs assuming a participation rate of 42%.

FIs website were checked and validated before it was added to the database. Figure 8 shows the selected criteria and summarizes the stratification method.

1st Stratification (FIs Sources)

FIs from the Bibliography

Google & Finance in Common

2nd Stratification (FIs officials website)

Global South regions

Funding Agriculture

3rd Stratification (FIs intervention type)

Local FIs

International FIs

33

Figure 8: Listing criteria of financial institutions in the final database

The full information about the database can be found in Annex B. 3.5.3. FIs Listing:

A spreadsheet with FI's data was listed in an excel file as described int the previous section. There are six sheets in this Excel file. First, we provide a summary of the general classification of each FI, and we explain the different colors and abbreviations used. A further four Excel sheets bearing the names respectively Africa, Asia, America, and Oceania were then created, holding FIs from the search engines and from the finance in common platform. Finally, the sixth Excel file contains the final sample with all the information required for data processing.

34

4. Survey Design and Conceptual Framework

4.1. Exploring other surveys

4.1.1. Survey with FIs

In the bibliographic reference, only an interview made by Röttger (2015) was conducted with FIs providing agricultural finance to smallholders' farmers in Uganda, Kenya, Benin, and Cameroon. The FIs involved are two commercial banks (CB), three microfinance Companies (MFCs) and three membership-based financial institutions (MBFI).

This survey⁷ contained primarily general information about FIs such as the country of origin, year of establishment, total assets, Gross loan portfolio, deposits, number of clients and branches. Then, the author has determined the percentage of the agriculture, loan portfolio of the total loan portfolio for the interviewed FIs and classified their loan product as a short-term loan (=12 months) and medium-term loans (=12 months).

Table 14: financial Institutions General Information

Financial Institutions General Information		· Country of Origin · Year Of establishment · Total assets · Gross loan Portfolio · Deposits · Number of clients · Number of branches	Source: Röttger (2015)

The author has identified the specific loan features for both short-term and medium-term loans such as Loan amount, Loan term, repayment schedule, collateral requirements and joint liability and other features. She has found that these FIs imply low entry barriers, more specifically for short-term loans, flexible repayment schedule, collateral requirements like group lending, and cash collateral and personal guarantors

Meanwhile, the authors of KIT and IIRR (2010) have listed the key characteristics used to depict financial product. These are amount and period of the loan, the disbursement, the way of repayment, the interest rate, and transactions cost.

7 Source: UMM Thematic Paper Assessed: 21 July 2021

See https://www.e-

mfp.eu/sites/default/files/resources/2015/03/UMM%20Thematic%20paper%202014 frankfurt final.pdf

35

Following the review of the two references, the specific loan features are summarized in the following Table.

Table 15: Specific loan features

Loan Features	· Loan amount · Loan term · Repayment schedule · Collateral requirements · Liability · Other features
Source: Röttger (2015) & KIT and IIRR (2010)

Through these previous studies, we can conclude that there is consistent evidence that FIs play not just an important role in financing the agricultural VCs but engage with chain actors to develop new financing models, mitigate risks more effectively, and lower transactions costs. Nonetheless, there is a lack of robust research on the paramount role of FIs in implementing approaches such as AVCF, and the mechanisms used by FIs concerning credit screening, scoring, rating tools, and portfolio monitoring practices, which should be improved, developed, and replicated.

4.1.2. Survey with farmers (World Bank & CGAP)

The World bank⁸ in cooperation with the Consultative Group to Assist the Poorest (CGAP) conducted a nationally representative survey of smallholder farmers in Bangladesh, Ivory Coast, Mozambique, Nigeria, Tanzania, and Uganda. The survey investigated the farming and non-farming activities, financial practices, and interests, as well as challenges and aspirations of smallholder families at the national level in six countries. Besides, this survey in each country has a nationally representative sample of about 3,000 families who indicated that agriculture makes a significant contribution to their household's livelihoods, income, or consumption. In total, more than 300,000 data points on the financial lives of smallholder families were collected from household surveys (World Bank and CGAP, 2017).

This Smallholder Families Data Hub provide data about the Poverty Status, Income Sources, Regular Expenses, Household Segmentation, Crops and Livestock, Saving for Agriculture, Agricultural Markets, Financial Inclusion, Financial Tools, and Trust in Financial Services Providers of small-scale farmers. This table summarizes the focal points of the above findings. This research provides more detail on these focal points and profiles of the four main segments

8 Source: CGAP Assessed: 31 July 2021

See https://www.cgap.org/small holders data portal/

36

of smallholder households in the six developing countries. These findings are designed to identify opportunities for FIs, government, and NGOs to better adapt the financial tools and meet the needs of smallholder families

Table 16: focal points of the World Bank Survey

Poverty Status	Income Sources	Regular Expenses	Household Segmentation	Crops and Livestock
ü Less than $ 1.25/day ü Between $ 1.25 and $2.50/day ü Above poverty	ü Crops sale and Livestock ü Business in Retail and Manufacturing ü Salary from regular jobs ü Rent Land ü Remittances ü Occasional job	ü Pay utility bills ü Pay a school fee ü Deposit money ü Send/receive money	ü Farming for sustenance ü Battling the elements ü Options for growth ü Strategic agricultural entrepreneurship	ü Growing crops ü Raising animals		Saving for Agriculture
Agricultural Markets	Financial Inclusion	Financial Tools	Trust in FIs	ü Very important ü Somewhat important ü Not important
ü Wholesaler ü Retailer ü Direct to public ü Processor ü Other ü Government Agency ü Cooperative	ü Included ü Not included	ü Bank ü Mobile Money ü MFI ü Cooperative ü Semi-formal savings ü Post office bank	ü Banks ü Bank agents ü MFI ü Mobile money providers ü Savings groups

The goals of this Smallholder Survey are first, generation of a clear picture of the smallholder sector (household demographics, agricultural profile, poverty status, market access). Second, a segmentation of smallholder households in the six countries according to the most compelling variables (at least these focal points) that emerge. Third, a characterization of the demand for financial services in each segment, considering the needs, attitudes and perceptions of clients related to both agricultural and financial services. Fourth, an explanation of the financial needs of each segment and how they can be met, with both informal and formal service s, and where there may be promising opportunities to add value (Anderson, 2017).

37

4.1.3. Integrated Financing for Value Chains (WOCCU)

The publication⁹ of the World Council of Credit Unions illustrates how credit unions can fill the agricultural lending hole and create market links. It explores how these unions can provide value chain finance to improve productivity, promote economic growth, and ensure food security among small farmers (WOCCU, 2009). Moreover, a large part of their success in value chain finance is due to their strong community ties, presence in rural areas, and experience lending to low-income individuals and small businesses. Additionally, they developed new risk management strategies for lending to these producers.

At each stage of agricultural lending, these credit unions have developed 10 ways to manage risk. The steps involved are:

1. Ensure demand for crops on the market

2. Establish proper policies and procedures

3. Evaluate real financing needs.

4. Assuring individual loans with appropriate guarantees

5. Diversify the portfolio of loans

6. Adapt loan terms based on crop seasons.

7. Using vouchers to distribute loans

8. Encourage farmers to diversify their crops

9. Monitoring crop performance

10. Using the Credit Union to receive payments

This publication argued that value chain finance methodology can be relatively inexpensive to implement and can be adapted to different contexts, products, and environments. However, the value chain must be managed by someone who can bring together, facilitating, and ensuring that business relations are maintained to result in an efficient value chain that meets all participants' needs.

4.1.4. Survey on national development bank (World Bank Group)

The paper summarizes the main findings from the 2017 survey of the national Development Banks conducted worldwide. Sixty-four DBs were surveyed from around the world, mainly from middle-income countries. Among the conclusions in this report is that although DBs tend to be small in terms of assets, governments use them to provide financial services to sectors or regions that are not well served by private intermediaries. Additionally, development banks operate in multiple economic sectors, contribute to global development agendas, and support

9 Source : WOCCU (2009) assessed : 08/10/2021

https://www.woccu.org/documents/value chain techguide

38

the private sector within their jurisdictions. Several institutions that participated in this survey noted that they face various challenges, including enhancing their risk management capacity and adapting better monitoring and evaluation frameworks (WBG & WFDFI, 2018).

There are 16 sections in this questionnaire with 138 questions. We see the use of a variety of question types including yes/no, multiple choice, and open-ended questions. These 16 divisions in this survey¹⁰ are as follows:

1) General information 9) Pricing and subsidies

2) Mandate 10) Profitability, asset quality, and efficiency

3) Size 11) Corporate governance

4) Loan portfolio and guarantees 12) Transparency and disclosure

5) Countercyclical Role 13) Prudential regulation and supervision

6) Funding 14) Restructuring

7) Business model 15) Monitoring and evaluation practices

8) Products and services 16) Challenges

Source: World bank Group, 2018

4.2. Credit Scoring for Agricultural Loans

For the stability and the profitability of FIs, loan contracts must perform well, and the screening of loan applications from agribusinesses should minimize credit risks and reduce TC. In this regard, a credit analysis must be done by FIs before making a decision on credit accreditation as part of the examination process. This credit analysis takes into consideration the assessment of the financial backgrounds and history of the applicant. Therefore, a good credit management assists FIs in loan pricing, determining the amount of credit, reducing the default risk, increasing debt repayment, and predicting the credit worthiness of borrowers (Limsombunchai, Gan, & Lee, 2005). Even more so, credit scoring not only assist FIs in loan approval, but also on loan monitoring and assessment of loan portfolio risks (Turvey & Brown, 1999).

According to Plata and Nartea (1998), credit analysis is the first step in a loan request. This covers the determination of the candidate's strength, estimating the probability of failure and reducing to an acceptable level the risk of nonpayement. In this context, several authors found that this method based on the evaluation of a loan officer sound to be not efficient (Crook, 1996; Glassman & Wilkins, 1997).

10 WBG & WFDFI (2018) Assessed 09/10/2021

https://documents1.worldbank.org/curated/en/977821525438071799/pdf/2017-Survey-of-National-development-banks.pdf

39

A study made by Limsombunchai et al. (2005) on the estimation of lending descision for agricultural loans has enumerated several factors used by FIs on credit scoring. These factors include borrowers' liquidity, profitability, solvency, and repayment capacity. These variables can be inferred directly from the applicant's financial status. In addition to that, moneylenders consider the farmer's personal attributes, entreprise type, region, and many other factors mentioned in Section 2.5 (Table 3). On top of that, the relationship between the bank (lender) and the borrower influence the availabilty and cost of the loan (Petersen & Rajan, 1994; Harhoff & Korting, 1998).

Limsombunchai et al. (2005) have developped a logistic model for the credit scoring which is a function¹¹ of the borrower characteristics, credit risk, proxies, relationship indicators and dummy variables. This variables excluded are the estimation of borrower's liquidity and repayment capacity. The credit scoring models is as follows:

Lending decision = f (Borrower characteristics, Credit risk proxies, Relationship indicators, Dummy variables)

The variables of the function are summarized in the following table as well as those excluded from the function.

Table 17: Lending decision variables

Borrower characteristics	Credit risk	Relationship indicators	Dummy variables	Excluded variables
I Assets I Age I Education	I Collateral I Return on assets I Leverage ratio I Capital turnover ratio	I Borrowing from others I Duration	I Province I Farm type I Loan Type I Loan size I Lending year	I Borrower's liquidity I Repayment capacity
Source: (Limsombunchai, Gan, & Lee, 2005, p.1199)

According to the authors, these variables are going to be rated by (+) or (-) in order to determine the lending decision. For example assets, age, collateral, capital turnover ratio, and duration are positevely related to the chance of a good credit. However leverage ratio and borrwing from other are negatively related to the chance of a good credit (Limsombunchai, Gan, & Lee, 2005).

On the basis of the case studies showing the practice of the AVCF approach in the following value chain; potato in Peru, milk and organic quinoa in Bolivia, cotton in Tanzania, sal leaf in India, rice in Rwanda, honey and tea in Kenya, soybean in Ethiopia, coffe in Nicaragua. The

11 Source: Limsombunchai, Gan, & Lee, 2005, p. 1199 Assessed : 22 July 2021 See https://thescipub.com/pdf/ajassp.2005.1198.1205.pdf

40

41

authors of KIT and IIRR (2010) have established the key factors to depict financial products summed up in the following table¹².

Table 18: Loan accreditation Characteristics

Product and financial flows	Risk management		Information flows
ü Purpose ü Amount ü Period ü Disbursement ü Repayment ü Interest Rate ü Transactions costs		Securitization Liability		Information required to apply Information required during season Time lag between application and payment
Source: (KIT and IIRR, 2010)

On this background of the last reference, we can determine the factors and variables for Credit screening, scoring and monitoring for Agricultural loans (Limsombunchai, Gan, & Lee, 2005; KIT and IIRR, 2010). This will be used in the following section to determine the questions for part D of the survey

4.3. Financial instruments employed by FIs

There are several ways to categorize the terms and describe the different financial products and instruments, e.g., Wenner (2006) categorizes these terms by product financing, receivables financing, physical asset collateralization, risk mitigation and structured enhancements. However, Miller and Jones (2010) in their books: «Agricultural Value Chain Finance Tools and Lessons» have categorized these terms and conditions differently. They classified the 16 instruments by 5 financial products as shown in the following Table¹³.

Table 19: AVCF Instrument

Product financing		Receivable's financing		Physical asset collateralization		Risk mitigation products		Financial enhancements
	Trader credit		Trade		Warehouse		Insurance		Securitization
	Input supplier		receivables		receipts		Forward		instruments
	credit		finance		Repurchase		contracts		Loan
	Marketing		Factoring		agreements		Futures		guarantees
	credit company		Forfaiting		Financial leasing				Joint venture finance
	Lead firm financing
Source: (Miller & Jones, 2010, p. 56-57)

12 Source: KIT and IIRR, 2010

Assessed: 25 July 2021

See https://www.kit.nl/wp-content/uploads/2018/08/1610 chainfinance-d8.1.pdf

13 Source: Miller & Jones, 2010

Assessed: 22 July 2021

See http://www.fao.org/3/i0846e/i0846e.pdf

The authors have mentioned that the use of these terms' changes according to the countries, moreover it can have other appellation and application of the instruments beside other legal terms. This table includes the traditional forms of credit and other more sophisticated and complex forms, and therefore some instruments may not be applicable to small-scale farmers, but on the other hand applicable for agro-industries and wholesalers. These sophisticated high-level instruments can in any case stabilize prices and reduce financing risk (Miller & Jones, 2010).

4.4. Survey design for FIs officials

After reviewing the literature and exploring several references, considering the content of other surveys, and the revision of determinants of agriculture credit, credit scoring mechanisms and the different instruments presented early, we designed the online questionnaire for bank officials Annex C. This was based on the consideration of other surveys oriented for both farmers and FIs. To capture the complexity of the FIs survey, the online questionnaire consisted of five sections already announced in part 3.4 [structure of the survey]. For each FIs, an online questionnaire was administered to the CEO or head of office or director, to collect the information.

· General Information

· Agricultural finance

Section

Section

· Financial product

Section

· Economic Information

· Credit scoring

Section

Section

most
financial
instrume
nt
employed

Loans
approval

Data from the
five sections
generates
relevant
information
about FIs agenda

Major FIs
informati
on:
Country
& Type

way of
using the
concept
AVCF

business
of FIs;
branches

& Clients

Figure 9: Financial institutions Questionnaire components

42

The FIs questionnaire collect first General & Economic Information, on whether each FIs contributes to VCF or participate to the AVCF. The information was later used to identify how FIs apply the approach AVCF in the Global South. Information on assets and Branches were also collected to derive the socioeconomic status of the FIs and the priority of financing the agricultural sector. In addition, credit accreditation criteria, financial products, and instrument employed were also enquired to find out how the FIs is funding rural producers. The below figure summarizes the sections of the survey.

All 5 sections were translated in Arabic, French, and Spanish, then pretested in August 2021. After the pretest, debriefing Sessions were held with university staff and Experts to modify the online questionnaire based in the observations and the Feedback from the pretest. Next the questionnaires were finalized and a script in the platform Qualtrics were programmed to facilitate data collection. The script was tested and validated by the Supervisors.

The following is in an explanation of theses 5 sections building on other studies. This methodology was designed to answer a few questions about the activities of FIs in the agricultural sector in the Global south.

4.4.1. General Information

This part will be devoted to general information about the institute and the position of the respondents. This first section permits the understanding and the segmentation of FIs according to several key characteristics (e.g., demographics, institutional type, experience in the agricultural field) and what type of FIs emerge? What is the position of the person who responded to the survey? How many years has he or she worked in agricultural finance?

This first part will allow the classification of FIs and then draw conclusions based on the latter segments.

4.4.2. Economic information

This second section will collect economic information from FIs such as (Gross loan Portfolio, No of Branches, and No. of Clients) and will determine the importance of the agricultural sector among the other branches in terms of budget from capital granted for it and in terms of customers. Additionally, this part highlights the attitudes and perceptions of FIs. How they perceive their agricultural activities? What is the priority sector? What is the penetration rate?

4.4.3. Credit screening, scoring, and monitoring for Agricultural loans

This part will determine the characteristics of credit accreditation and evaluate credit screening, scoring, and monitoring mechanisms for agriculture loans. This will allow us to have statistics on the different key characteristics on lending decision and to determine the differences in

43

consideration of this criteria between the different countries and continents. Which criteria does each segment of FIs demand, from the perspectives of both rural producers (age, education, marital status, agricultural experience, etc.) and financial products (e.g., amount, term, repayment schedule, collateral)?

4.4.4. Agricultural finance within value chains

This section will investigate the FIs-official of the financial institute on the knowledge and application of the AVCF. This will help us determine the percentage of FIs that finance agriculture with this approach and understand the opportunities to improve the financial inclusion for small-scale farmers and agribusinesses in the one hand, and in the other hand mitigate risk and reduce transaction cost for the FIs. Where are the opportunities to add value with the approach AVCF?

4.4.5. Financial product & Instrument employed

Based on the instruments determined by Miller and Jones (2010), this last section will investigate the respondents on the instruments used. What formal suite of financial mechanisms does each segment of FIs currently use? These 16 instruments were classified into 5 groups: Product financing, Receivable's financing, Physical asset collateralization, Risk mitigation products, and financial enhancement as shown in the figure below.

Trader
credit

Input
supplier

Marketing
credit
company

Lead firm
financing

Trade
receivables
finance

Factoring

Forfaiting

Warehouse receipts

Repurchase agreements

Financial
leasing

Insurance

Forward
contracts

Futures

Securitizatio

n

instruments

Loan

guarantees

Joint
Venture
finance

Product financing

Receivables financing

Physical asset collateralizatio

n

Risk mitigation products

Financial
enhancements

Figure 10: List of instruments enquired during the survey

4.5. Overview of the online questionnaire

During the bibliographic study, few research, discussion, and paper have identified a gap in the contribution of FIs in the agricultural credit (Meyer R. L., 2007; Yadav & Sharma, 2015),

44

evoked the needs, desires, and perceptions of FIs. This master's thesis involved defining the end objective by doing the following:

§ Drawing conclusions from existing survey

§ Considering the objectives and needs of FIs

§ Accounting for FIs-officials Feedback

§ Learning from the ongoing financial agendas in each continent.

Using these building blocks, a framework for the online survey was developed to be shared with FIs and to capture all relevant elements of the AVCF. The framework thus comprised five main sections as previously announced is summarized by the following figure:

Economic Information

General Information

Agricultural finance

within value chains

Credit scoring

meschanisms
· Assess Borrower Characteristics

Instrument

· Identify and profile FIs

· Determine the agricultural experience

· Measure penetration rate

· Find out the priority sector

· Discover Loan accreditation criteria

· Examine the practices of the application of the AVCF

· Determine the potential products and services

Figure 11: Questionnaire Framework

5. Analysis and Results

5.1. Descriptive Analysis:

5.1.1. Geographic distribution:

When examining the number of financial institutions, Africa dominates the database, accounting for 47% of all institutions. The Asian and South American FIs have a relatively low proportion of their cases studied in this database, at 24% and 20%, respectively. The rest of the database is composed of international financial institutions (IFI). The global scope of the IFIs enables them to operate in different developing countries and help promote agriculture development, sustainable resource management, and resilience. An overview of the database can be found in the following Table 20.

Table 20: Classification of financial institutions by continent

Continent

Africa

South America

Asia

Global

Total

No of FIs	Share
67	47%
35	24%
29	20%
13	9 %
	144

45

5.1.2. Distribution by institutional type:

Based on the literature review on agricultural finance, our database suggests that they account for less than ten percent of total FIs. Most agricultural credit is provided by cooperatives, microfinance institutions, and partly by commercial banks (Table 20). Financial institutions in this database were classified according to the descriptions found on their own websites, predominantly in the rubric "about us" or "history". To gain an understanding of the most important FIs, commercial bank as described in the literature are financial institutions that accept deposits, provide services of checking accounts, offer various loans, and offer products like personal loans and business loans for both individuals and small businesses. However, cooperative are organizations where many small farms work together to produce and sell crops together. The cooperatives also provide the farmers with short-term agricultural loans which allow them to carry out many different agricultural and farming activities in the collective. In addition, a significant share of this database consists of MFI which is an alternative banking service that is available to unemployed or low-income individuals who would otherwise have no access to financial services, and development bank which provides capital to create productive investment opportunities, often in conjunction with technical assistance. Finally, Islamic bank is a financial or banking entity operating under shariah (Islamic law) and making

46

profits through equity participation, which entails the borrower giving the bank a share of their profits instead of paying interest.

In light of the observations and case studies in this database, the cooperatives and MFIs do not deliver enough information to provide a meaningful sample or claim that agriculture financing is primarily provided by cooperatives. In some cases, cooperative groups are organized and monitored by non-governmental organizations, in others by international financial institutions and international funds, and in yet others, commercial banks lend to the cooperatives that then distribute the funds among the farms according to various financial institutions' websites which provide information on this. Additionally, there are cooperatives that collect farm savings and distribute them as needed in an organized and priority manner to their registered members.

Cooperation has an important role to play in increasing agriculture credit availability, connecting commercial financial institutions, and ensuring small-scale farmers' financial inclusion.

Table 21: Classification of financial institutions by institutional type

Institutional type Number %

Commercial Bank 48 33%

Cooperative 47 33%

MFI 21 15%

Development Bank 9 6%

Islamic Bank 5 3%

International FI 3 2%

International Fund 3 2%

Agricultural Bank 2 1%

Credit union 1 1%

Insurance company 1 1%

NGOs 2 1%

State development bank 2 1%

Total 144

5.1.3. Foundation Year:

As shown in Table 22, according to this database, 42% of the financial institutions have been established since 1990. The rest of the FIs are over 50 years old. Furthermore, when examining the data, there are no conclusive evidence that financial institutions invest in agriculture after several years in the banking industry.

47

Table 22: classification of financial institutions by foundation year

Foundation Year	Frequency Percentage
Before 1950	19	13%
1950 - 1970	32	22%
1970 - 1990	33	23%
After 1990	60	42%
Total		144

5.1.4. Number of Branches

Based on analysis of the number of branches per FIs, we find that 29% have less than 10 branches, 44% are in the interval [10-100], and 22% have between 100 and 1000 branches. Eventually, only eight FIs in this database have more than 1000 branches in the origin country.

Table 23: Classification of financial institutions according to the number of branches

Range	Africa	Number of Branches of FIs per Region America Asia Global Grand Total				Percentage
< 10	23	12	5	2	42	29%
10 - 100	31	14	13	5	63	44%
100 - 1000	12	7	7	5	31	22%
> 1000	1	2	4	1	8	6%
Grand Total	67	35	29	13	144	100%

According to Table 23, more than half of the individuals in this simple with more than 1000 branches are from Asia. Several factors can explain this, including the high concentration of population in India.

5.1.5. Agricultural loans

A total of 84 Farmer Loans (AL1) and 24 Agribusiness Credit (AL2) were found among the 144 FIs in this database. There are several important institutions contributing to this financial offer - mostly cooperatives. They hold the top position with 37 AL1 and 5 AL2 (for value chain players), followed by commercial banks with 16 loans for farmers and 14 for value chain players. Next come the microfinance institutions, which offer 15 agricultural loans and 5 chain loans. Both types of loans are provided by all agricultural banks.

As for the other institutional types, the agricultural loan offer is either minimal or absent since development banks or international funds provide these loans through other organizations or financial institutions including cooperatives and micro enterprises.

48

Table 24: Percentage of credit offered by type of financial institution

Type	Sum of AL1	% AL1	Sum of AL2	% AL2
Agricultural Bank	2	100%	1	50%
NGOs	2	100%	0	0%
International Fund	3	100%	0	0%
cooperative	37	79%	5	11%
MFI	15	71%	5	24%
International FI	2	67%	0	0%
State development bank	1	50%	0	0%
Development Bank	4	44%	3	33%
Islamic Bank	2	40%	0	0%
Commercial Bank	16	33%	14	29%
Credit union	0	0%	0	0%
Insurance company	0	0%	0	0%
Total	84		28

Table 24 shows that all agricultural banks, international funds, and NGOs offer credit to small-scale farmers in this database, although cooperatives and MFIs are more likely to use this form of credit. International funds are slightly less prominent in this database. In terms of agribusiness loans (AL2), only 50% of agricultural banks offer these services, 29% of commercial banks, 33% of development banks, and 24% of MFIs.

5.1.6. Gender Equality

For gender equality, the 144 FIs sample covered 22 Credits facility for women (G1), 15 Career development opportunities to female staff (G2), and 43 Gender Programmes (G3).

Generally, other institutional types have little direct interaction with VCs and are more concerned with regional development. Because of this, these institutions almost have no gender-related offerings.

As a result of reviewing table 25, we can learn that 50% of agricultural banks and NGOs offer women-credit, 50% of nongovernmental organizations and 50% of state developmental banks provide gender equal employment opportunities and more than 50% of international funds, Islamic banks, and nongovernmental organizations provide assistance and programs to women.

49

Table 25: Percentage of gender equality program offered by type of financial institution

Type	Sum of G1	%G1	Sum of G2	%G2	Sum of G3	%G3
Agricultural Bank	1	50%	0	0%	0	0%
NGO	1	50%	1	50%	1	50%
International FI	1	33%	1	33%	1	33%
International Fund	1	33%	0	0%	2	67%
MFI	6	29%	4	19%	6	29%
Development Bank	2	22%	0	0%	3	33%
Commercial Bank	6	13%	2	4%	17	35%
cooperative	4	9%	6	13%	10	21%
Credit union	0	0%	0	0%	0	0%
Insurance company	0	0%	0	0%	0	0%
Islamic Bank	0	0%	0	0%	3	60%
State development bank	0	0%	1	50%	0	0%
Total	22	15%	15	10%	43	30%

5.1.7. Digital Solutions

In terms of digital solutions, we found that 60 FIs out of 144 offer Online Banking (DS1), 19 FIs offer E-Products (DS2), and 18 FIs offer Online Loan Applications (DS3). First place goes to the commercial banks with 37 DS1, 10 DS2, and 11 DS3. Cooperatives rank second, followed by MFIs and Islamic banks. The other typical institutions do not have a strong focus on digital solutions, whose offers are very limited.

Table 26: Percentage of digital solutions offered by type of financial institution

Type	Sum of DS1	% DS1	Sum of DS2	%DS2	Sum of DS3	%DS3
Islamic Bank	4	80%	1	20%	2	40%
Commercial Bank	37	77%	10	21%	11	23%
NGO	1	50%	0	0%	0	0%
MFI	5	24%	4	19%	1	5%
cooperative	10	21%	4	9%	3	6%
Credit union	0	0%	0	0%	0	0%
Development Bank	0	0%	0	0%	1	11%
Insurance company	0	0%	0	0%	0	0%
International FI	0	0%	0	0%	0	0%
International Fund	0	0%	0	0%	0	0%
Agricultural Bank	0	0%	0	0%	0	0%
State development bank	0	0%	0	0%	0	0%
Total	60	42%	19	13%	18	13%

The table shows that more than two thirds of commercial banks and Islamic banks offer online banking, that about 20% of MFIs, Islamic banks and commercial banks offer e-products, and that 40% of Islamic banks and 23% of commercial banks offer online loan applications.

50

5.2. Cluster Analysis

Our Data Analytics work often requires a large dataset (144 observations of 12 variables), which are nevertheless similar to one another. Therefore, together with similar observations within a cluster, we may need to organize them in a few clusters. These Clustering techniques are used in such situations to identify segments within the data. The similarity and the difference between data observations (e.g., geographical variables and institutional types) can also be understood mathematically using distance metrics, and different segmentation solutions can then be proposed. This section will consider 2 types of clustering methods. We will first consider k-means clustering, then Hierarchical Clustering.

5.2.1. Confirm Data:

Clustering can be done even when data are not metric; however, many statistical methods currently used for clustering require that the data themselves are metric: not only must the data be numbers themselves, but the numbers also have a numerical significance, as explained in section 3.3.4. This is required to compute distances between observations, and distances are usually computed only using metric data. However, it is possible to define distances also for non-metric data (e.g., continent, institutional type, agricultural credit, etc.). In our case, the data are metric, so the next step is to evaluate the descriptive statistics. Before moving forward, we check the descriptive statistics of the data to understand it better.

Table 27: Descriptive statistics of the dataset

	Mean	Median	Standard Deviation	Sample Variance	Range	Minimum	Maximum
C	0.96	1	1.04	1.08	3	0	3
T	3.44	2	3.16	9.95	11	0	11
F	1976.50	1984	30.36	921.78	180	1838	2018
B	215.08	30.5	661.57	437670.13	5653	0	5653
AL1	0.58	1	0.49	0.24	1	0	1
AL2	0.19	0	0.40	0.16	1	0	1
G1	0.15	0	0.36	0.13	1	0	1
G2	0.10	0	0.31	0.09	1	0	1
G3	0.30	0	0.46	0.21	1	0	1
DS1	0.42	0	0.49	0.24	1	0	1
DS2	0.13	0	0.34	0.12	1	0	1
DS3	0.13	0	0.33	0.11	1	0	1

51

5.2.2. Scale the data

The fact that some variables have very different ranges and scales can often lead to problems: the results can be largely shaped by a few large values. It is recommended to standardize the data, for instance by making some raw attributes have, for example, a mean of zero and a standard deviation of one, or to scale them between 0 and 1 to avoid such issues. After examining the statistical description of the data, and before proceeding with the clustering analysis, this step¹⁴ was carried out with the function(r). This step of scaling is achieved by dividing the values in every column by the corresponding 'scale' value from the argument if the value is numeric. If the value is not numeric, then the values are divided by the standard deviation or root-mean-square.

In our case, each variable has a mean of zero and a standard deviation of one, as expected. A copy of the R script is attached in [Annex D].

5.2.3. Select segmentation variables

The choices of which variables to use for clustering are crucial decisions that will greatly influence the clustering solution. Every variable chosen should have a significant impact on clustering. This exploratory research provides an idea of which factors may differentiate regions, type, etc. In our case, we can use the 7 categories of variables (Continent, Institutional Type, Foundation, Number of branches, Agricultural credit, Gender equality, and Digital Solutions) for segmentation, and the remaining 3 (Total assets, Total equity, and Gross loan Portfolio) cannot be considered as the data are incomplete.

5.2.4. Define similarity measure:

In clustering and segmentation, objects are related in some way. Most statistical methods for clustering include a measure of distance. For example, the Euclidean distance and the Manhattan distance are common measures. We will use K-means here, which is based implicitly on pairwise Euclidean distances between data points, and where the sum of squared deviations from centroid is equal to the sum of pairwise Euclidean distances divided by the number of points.

5.2.5. Number of clusters

Clustering and segmentation can be done statistically in a number of ways, so it might make sense to try several approaches before finally selecting one that is statistically robust, interpretable, and actionable - among other aspects.

14 In the R script, this step is under the heading # scale data

52

Throughout this study, we will utilize two widely used methods: the K-means Clustering Method, and the Hierarchical Clustering Method. Each method requires that we decide on how to measure distance/similarity between our observations. However, one difference to highlight is that K-means requires the user to specify the number of segments to create, whereas Hierarchical Clustering does not.

First, let's look at the K-means method. WSSplot will be used to determine the optimum number of clusters. Below is the result of this function for the data scaled points.

Figure 12: Determination of the optimum number of clusters

5.2.6. K-means Clustering Method

According to one of the most common typical cluster methods, the Euclidean distance of the observation can be used as a criterion for determining the optimum number of clusters. This option specifies how the distance between each observation point and the cluster center is calculated. This case will utilize the default Euclidean distance for measuring the distance between two points by using the length of the straight line between them.

The 144 provided observation points will be clustered in 2, 4, 5 and 8 clusters respectively. The other algorithm parameters will remain the same except for the number of clusters. By segmenting the data, it will be possible to develop (potentially more refined) segment-specific insights.

53

Data scaled Grouping in 2 Clusters Data scaled Grouping in 4 Clusters

Data scaled Grouping in 5 Clusters Data scaled Grouping in 8 Clusters

Figure 13: Grouping Data scaled in different clusters

To visualize the relationship between the category's variables (C, T, F, B, AL1, AL2, G1, G2, G3, DS1, DS2, DS3), we demonstrate this with the function fviz_cluster as shown in the Figure 14 below.

Figure 14: Clusters Visualization

54

Using the function $cluster we can determine to which cluster belong each of these observations. Then, all data observations were clustered in four membership groups

Table 28: Cluster membership IDs using K means method

Cluster no.	Observations Number							Total
1	3 18 35 36 42 48 49 50 62 63 76 84 111 125							14
2	2 7 17 20 26	37 67 75	82 88 114 120 134 135 137					15
3	10 41 43 47	57 59 64	72 73 74 78 90 92 94 101 102 103	104	117	124	126	29
	127 132 133	139 140	142 143 145
4	4 12 21 22 23 30 32 34 39 40 52 53 54 56 60 69 70 80 81							19
5	1 5 6 8 9 11	13 14 15	16 19 24 25 27 28 29 31 33 38 44 45 46		51	55 58 61		67
	65 66 68 71	77 79 83	85 86 87 89 91 93 95 96 97 98 99	100	105	106	107
	108 109 110	112 113	115 116 118 119 121 122 123 128	129	130	131	136
	138 141

A second way to identify if there is a relationship is to use Hierarchical Clustering (next Section).

5.2.7. Hierarchical Clustering Method

Hierarchical clustering figure out how many segments our data has. Additionally, this method gives a visual representation of how the data might be clustered. It produces a plot called the Dendrogram often helpful for visualization. For instance, in this case the dendrogram was based on the Euclidean distance metric and R's ward.D hierarchical clustering option is as follows:

Figure 15: Hierarchical Clustering

As shown in the Dendrogram, clustering is accomplished by aggregating observations by
narrowing them down to pairwise observations closest to each other and merging smaller

55

groups into larger groups based on which groups are closest in proximity. Our data are merged eventually into one segment. According to the heights of the tree branches, the clusters that have merged at that level of the tree are very different one from the other. The longer lines indicate the differences between the clusters below. This method combines the closest data points/groups first, then the farthest, resulting in increased branches heights when you traverse the tree from the ends to the roots.

Even with many observations, dendrograms can be useful for segmentation because, in general, branches grow logarithmically with data numbers. By analyzing the dendrograms, we can understand our data and the segments that exist in it better in practice. This Hierarchical Clustering found 4-segment solution (using Euclidean distance and hclust with option ward.D). All data observations were clustered in four membership groups.

Table 29: Cluster membership IDs using Hierarchical method

	Cluster Membership 1		Cluster Membership 2						Cluster Membership 3	Cluster Membership 4
Observation	7 75	137 82 37 67	60	52	56	4	39	40	84 111 48 36 63	86 112 121 128
Number	2 17	134 26 120	54	88	70	53	80	22	125 30 32 27 69	129 107 109 123
	133		34	81	21	12	23		76 103 105 94	113 131 98 108
									102 126 90 92 47	130 93 1 66 83 87
									78 133 104 127	31 28 29 16 51 58
									73 74 139 101	95 77 85 33 5 46
									132 72 124 42	116 99 110 97
									143 144 42 64	106 122 19 119
									117 15 6 62 18 49	138 140 136 141
									50 3 35 44 65 13	25 9 11 100 115
									55 71 96 57 43 59	91 61 79 10 8 45
										20 114 118 68 89
										14 24
Total	12		17						53	59

5.2.8. Selected method and number of clusters

The numbers above represent the clusters that our observations belong to when we use K-means method for 5 clusters and hierarchical method for 4 clusters, for the same total number of observations. Because we use different methods, we do not need the observations to be in the same clusters, nor do the segment profiles that we will find next (A comparison of IDs between both methods was done using four clusters). However, one feature of statistically robust Clustering is that the observations are segmented according to similar characteristics regardless of the methodology used. In other words, the segments' profiles should not vary too

56

much when we use different approaches or variations of the data. Therefore, the K-means clustering method will be our main focus. Using the WSSplot method (section 5.2.5), it is possible to determine the point where the kink in the curve may occur so that an optimum number of clusters can be determined. Another method¹⁵ is to run K-means Clustering from 1 to 10 and use the function betweenss_totss to compare and determine the best number of numbers.

By crossing the hierarchy downwards, we can see that in the hierarchical method we will get different clusters, starting with 4 clusters at the height of 15, 8 clusters at the height of 10, 33 clusters at the height of 5, and even more clusters as the height becomes lower (figure 15). We will focus on K-means Clustering to avoid this unnecessarily large number of clusters. The difference between these two methods is that with K-means you can test, select, and verify the optimum number of clusters. However, if you choose a minimum height, when using the hierarchical method, the number of clusters may appear to be excessive.

Moreover, it is found that the 5 segments found are relatively resistant to changes in data subsets and clustering methods. In general, the observations are grouped in the same clusters with no major changes. Segmentation is judged by the robustness of both the statistical characteristics as well as many qualitative criteria: institutional type, agricultural credit, gender, and digital solutions.

This segmentation was validated using silhouette coefficient with the function silhouette to determine the robustness of our clustering. In this case, the silhouette width is understood as follows:

· Si > 0 indicates that the observations are well clustered.

· Si < 0 indicates that the observation was placed in the wrong cluster.

· Si = 0 indicates that the observation is in between two clusters

In this silhouette plot, there is no negative silhouette width, and the maximum value is greater than 0.14, indicating that our clustering using five groups is good. The method of mounting the clusters is always more reliable if the ratio is so close to 0.5. If, however, this method shows a significant result, the size of the clusters and the individuals differ from those determined by the previous method.

15 In the R script, this step is under the heading # Choosing K

Figure 16: Validation of the number of clusters

57

5.2.9. Extracting Results

Having chosen five clusters, we want to learn more about who the financial institutions in these clusters are and how the segments can be interpreted. The data within each of the selected segments must be visualized and understood for this purpose. As a result, we can extract the clusters from the data and add them to our initial data to perform some descriptive analysis. To analyze each cluster, we will name it according to the most significant pattern, as shown below.

Cluster Number Average Foundation Cluster Name

1 1948 Value chain oriented FIs

2 1967 Gender Staff FIs

3 1975 Farmer credit provider FIs

4 1990 Innovative digital newcomers

5 1982 Traditional Banking approaches

Table 30: Nomination of the FIs groups

58

# Value chain oriented FI:

This cluster has the highest average age in the sample and most financial institutions are African, with 14% coming from each Asia and South America. In addition, this sample consists of mainly commercial banks with a 68% share, 7% are cooperatives, and the same share is held by development banks. Located in the cluster are the oldest average foundation years 1948 and the most average number of agencies, 517.

There is a high rate of agri-business and SME lending among these financial institutions, which funds actors in the value chain with an 86% percentage., therefore the value chain finance specialty in this segment. Moreover, 57% of the institutions in the sample provide agricultural credit to farmers, ranked 3rd among the clusters. In addition, 71% of these firms emphasize gender programs, but only 7% of them provide credit facilities for women, and like 3 other groups have do not invest in women's Equality staff. Among these FIs, 71% offer online banking, and half offer online loan applications.

# Gender Staff FI:

In terms of geographic location, this segment is very diverse, with half of the FIs coming from Africa, 13% from Asia, 20 % IFI, and 20% from South America. Concerning institutional type, cooperatives account for 40% of the sample, followed by MFIs at 27%, commercial banks at 13%, and state development banks, NGOs, and international financial institutions at 7% each. In this cluster, the average foundation year is 1967, younger than the cluster before, and the average number of branches is 433.

As indicated above, the Gender Staff FI group is made up of FIs that 100% guarantee gender equality in the recruitment and selection of employees. Moreover, 47% of FIs offer credit facilities to women, and 13% offer gender-specific programmes such as workshops, events, and scholarships.

In relation to rural lending, 67% of FIs offer agricultural loans to small farmers, while 20% lend to businesses. This segment is less digitally oriented, as only 47% offer online banking and 20% offer E-products, while 7% offer online loan applications.

# Farmer credit provider FIs:

The segment consists of 29 institutions, 28% from South America, 24% are IFI, 24% from Asia, and 24% from Africa. Regarding institutional types, cooperatives make up 31%, commercial banks 21%, MFIs 14%, and Islamic banks and development banks each make up 10%. A few other typical institutions include NGOs, international FIs, and international funds. There are 215 branches in this group on average and the average foundation year is 1975.

59

As far as rural lending goes, this cluster is the most extreme, where 72% FIs offer credits to small-scale farmers while 0% offer loan to agribusinesses.

All financial institutions advocate gender equality and empower women through a range of programs. However, fewer than 7% of FIs offer credit to women, and there is no interest in offering employment opportunities to female staff.

In terms of digital solutions, half of the FIs provide online banking, however, only 7% offer online loan applications.

# Innovative digital newcomers

This cluster has 19 financial institutions, 79% of which are from Africa and 21% from Asia. 68 percent are commercial banks, 16 percent are cooperatives, 11 percent are microfinance institutions, and 5 percent are Islamic banks. Compared to other groups, this group has a very small number of branches, 95 on average. In this cluster, the average FI foundation year is 1990, making it the youngest.

There are the lowest bids for small farmers in this cluster, with 37% offering credit to farmers, 26% lending to value chain players, and gender equality isn't strongly addressed; 16% of small business loans go to women, and 11% promote gender programmes.

Additionally, this group is characterized by digitalization as 84% of FIs offer online banking services, as well as other electronic products like SMS Banking, Email Banking, and Mobile Loans. Moreover, 37% of the FIs in this segment offer Online Loan Applications.

# Traditional Banking approaches

This last group, with the largest number of members, was made up of 48% Africans, 34% South Americans, 19% Asians, and 4% global. The most dominant institution type is cooperative, representing 42% of total institutions, followed by commercial banks, 16% of microfinance institutions, 7% of development banks, 3% of agricultural banks, and 1% for credit unions, insurance companies, international funds, Islamic banks, and nonprofits. FIs in this traditional cluster have an average age of 39 and 137 branches on average.

This segment is the lower regarding bank services, where we found that half of this group offers agricultural credit for farmers, and 12% offer agribusiness credit. Gender-specific projects are not well supported by this group, only 13 provide women facility credit. Finally, the lower digital solutions offer with only 16% FIs offer online banking, 0% E-products, and 1% online loan application.

Below is a summary of each cluster's most important characteristics.

Table 31: Cluster's characteristics

60

Cluster Name

	Group 1 Value chain FIs		Group 2 Gender Staff FIs		Group 3 Farmer credit provider FIs		Group 4 Innovative digital newcomers		Group 5 Traditional Banking approaches
Region
Africa	71%		47%		24%		79%		42%
Asia	14%		13%		24%		21%		19%
South America	14%		20%		28%				34%
Global			20%		24%				4%
Institutional Type
Agricultural Bank									3%
Commercial Bank	86%		13%		21%		68%		22%
cooperative	7%		40%		31%		16%		42%
Credit union									1%
Development Bank	7%				10%				7%
Insurance company									1%
International FI			7%		3%
International Fund					7%				1%
Islamic Bank					10%		5%		1%
MFI			27%		14%		11%		16%
NGO			7%		3%				1%
state development bank			7%						1%

61

Average Foundation

	1948	1967	1975	1990	1982
Average Branches	517	433	215	95	137
Agricultural Credit
Farmer credit	57%	67%	72%	37%	57%
agri-business credit	86%	20%	0%	26%	12%
Gender Programmes
Credits facility for women opportunities to female staff	7% 0%	47% 100%	7% 0%	16% 0%	13% 0%
Gender Programmes	71%	13%	100%	11%	0%
Digital Solutions
Online Banking	71%	47%	55%	84%	16%
E-Products	0%	20%	0%	84%	0%
Online Loan	50%	7%	7%	37%	1%
Applications
Total FIs	14	15	29	19	67

6. Discussion

6.1. An information provider database

The database provides a snapshot of financial institutions engaged in agricultural sectors in the Global South, including African, Latin American, Asian, and international financial institutions. The different kinds of institutions such as commercial banks, cooperatives, development banks, Islamic banks, cooperatives, and credit unions, as well as international funds and institutions, each finance the agricultural sector differently. As a result, the database can be used to determine the similarities and differences among these institutions and understand how financial institutions, make available offers in the agricultural sector to meet the needs of rural clients.

There is a large amount of data that is available in the database from websites of financial institutions, starting with the geographical location and type of institution, which, in this case, do not appear to be correlated with the offer of agricultural loans in this case. Furthermore, the database listed the year the financial institute was founded to highlight the years of experience in banking on the one hand, and to determine whether this has any direct bearing on agriculture financing, but this is not the case in the present sample. As well, the number of branches of financial institutions and the relative importance of their dissemination in providing small-scale farmers with information and access to agricultural credit are not strongly linked in this database to agricultural credit offer. However, based on the literature review, it was indicated that rural areas were lacking an appropriate infrastructure, have no branches, or have a limited network which makes the rural fundraising situation more difficult (Langenbucher, 2005). Similarly, in other publications, low population densities, fewer bank branches and distant locations were cited as factors that reduce profitability of rural finance (IFC and World Bank, 2012).

Among the other important categories of information contained in this database are the availability of agricultural credit for farmers as input financing, labor payment, and harvest and storage, as well as credit for agribusiness and agricultural SMEs. As cited in the literature review of (ISF Advisors and Mastercard Foundation, 2019; UNCTAD, 2004), this category of variables shows a financing gap for the agricultural sector, despite the use of the approach VCF by many FIs. This sample contains only a little over half of the institutions that provide agricultural credit, and even fewer when it comes to financial institutions that provide financing to value chain participants, not even one fifth.

Regarding gender equality, observations show that not many gender programs are offered such as women's empowerment events, women's savings accounts, scholarships, workshops,

62

but when it comes to staff equality and credit for women, the available options are far from expectations.

In terms of digital solutions, the database indicates that increasingly financial institutions rely on online banking; on the other hand, e-banking products such as SMS, Email, and Mobile Loans are less and less common, and are only used in countries where smart phones are not yet widely available. As for online loan applications, FIs are experimenting with this new method, but the products available are very limited. According to the literature, FIs can enhance their ability to offer competitive products by strengthening the relationship between producers and buyers, which will lead to more innovative products (Kaufman, Wood, & Theyel, 2000).

Due to the inaccessibility of information on the part of farmers, and the low awareness of farmers' products by institutions, it is possible to argue that agricultural credit for small-scale farmers, as well as for value chain actors, exists in a weak and even random manner. Further, some financial institutions are reluctant to lend to the agricultural sector because of the lending risk as mentioned in the literature review (Herliana, Acip, Qorri, Qonita, & Nur, 2018).

6.2. Analysis of the clustering analysis

In fact, when assessing the factors that could facilitate the development of agricultural loans, the cluster analysis has revealed that regardless of geographic location, age of the institution, and number of branches, financial institutions with value chain lending programs offer more agricultural credit to farmers and support women with a range of programs. The majority of these FIs offer Online Banking, in addition to online loan application. In fact, this analysis has also demonstrated that when financial institutions promote gender equity strongly or programs that support women either directly or indirectly, they immediately offer moderate credit for women, agricultural credit for small farmers, and several digital solutions. A key finding of the study is that financial institutions offer customized and enhanced rural financial services that are in high demand and in line with gender issues and digital solutions. This was shown by the results of cluster analyses for the first, second, and third segments with the names respectively of Value Chain FIs, Gender Staff FIs, and Farmer Credit Provider FIs.

In contrast, when FIs provide low-level financial services and tend to use traditional banking methods, they do not promote gender equality or offer digital solutions, as shown in the cluster 5 (Traditional Banking approaches). Nevertheless, technology-oriented solutions are at the center of the most youthful cluster, with an average foundation year of 1990, hence the name Innovative digital newcomers.

63

Previously, the analysis looked at the variables of agricultural loans, women's equity, and digital solutions together. Keeping in mind that years of banking experience and number of branches may not be statistically significant. Findings indicate that commercial banks provide most of the credit to value chain approaches, while cooperatives are the most popular for agriculture loans, both with MFIs being gender equity-oriented organizations. As explained in the previous section, the dominance of commercial banks in the first segment Value chain oriented FIs and the fourth group Innovative digital newcomers in providing financial services and digital solutions may be explained by the fact that these types of institutions have the resources necessary to provide these services. While there are more individuals from these types of organizations may explain.

Based on table 19, commercial banks, cooperatives, and microfinance institutions represent 81 percent of our simple. However, this database shows all agricultural banks, international funds, and NGOs offering loans to small farmers, although cooperatives and MFIs are most likely to use it. International funds appear less frequently than agricultural banks. Agribusiness loans are the exclusive domain of 50% of agricultural banks, 29% of commercial banks, and 33% of development banks as well as 24% of microfinance institutions according to the 144 FIs sample

Additionally, table 21 shows that several FIs such as development bank are the least likely to provide agricultural loans. This could be because these institutions may be included as facilitators or brokers for several agricultural development projects. Furthermore, these institutions often provide financial aid through other financial institutions, and that is the major case for this sample of database according to the websites of FIs.

6.3. Limitations and further research needs

The evaluation of the database provides valuable information about the potential for diversifying financial offerings and introducing innovative products that cater to rural clients' needs.

First, this database was created based on an English literature review, so many financial institutions from North Africa, South America, and Asia were not considered.

Second, the structure of the database causes some information to be lost due to specific categories of data being classified into. Consequently, some of the information cannot be considered. This information relates to farmer credit for instance, where several FIs divide this product into several categories such as input loans, fertilizer loans, harvest credit, etc....Nevertheless, other FIs include this type of credit under one category of farmer loan. Furthermore, 4 FIs from the 144 in the sample work with service providers or collect information

64

about farmers for credit risk screening purposes. These factors are collected and used to determine whether a farmer is credit worthy or not. However, this information is not considered since these FIs represent only 3% of the sample. Therefore, a deeper analysis must be conducted of an expanded and more diverse set of financial institutions taking into account this important information.

Third, the data base also included several categories of variables, such as Total assets, Total equity, Gross loan portfolio, and number of clients. Since many websites did not have information on these variables, we did not take them into account. Consequently, these cannot be considered and high-level information, such as penetration rate, cannot be calculated at this point.

Fourth, the lack of credible or official websites of many financial institutions, which reduced the sample by 50 FIs. This is rather than their significant role in funding VC initiatives as described in the bibliographic study. In order to address these issues, an online survey should be conducted with FIs representatives to gather the maximum amount of information necessary for analysis. In this regard, an online questionnaire is crucial to analyze the way in which FIs currently fund the agriculture sector, including the approach AVCF. Having a clear understanding of how FIs apply the AVCF is crucial on the one hand! On the other hand, it is essential to provide an accurate depiction of the AVCF's features and ways it can improve access to finance for all value chain players.

Also, the foundation believes more research needs to be done on gender equality programs and their relationship to agricultural finance. A study is needed to determine whether promoting women's empowerment and gender equality can promote lending risk reduction and increase credit facilities in the agricultural sector. Though cluster analysis did not show a significant relationship between digital solutions and agricultural credit, evaluating how digital tools can be used to increase farmers' awareness of credit offers by financial institutions is a high priority.

A weakness of the analysis is the neglect of some variables including geographical location (G), institutional type (T), and foundation year (F), and the number of branches (B) due to the use of the binary method of accreditation for the qualitative variables. Thus, after duplicating the agricultural loan variables into two variables (AL1 and AL2), gender equality into three variables (G1, G2, G3), and digital solutions into three variables (DS1, DS2, DS3), we arrived at 12 variables in total. As a result, cluster analyses were influenced by inequalities in importance, of which eight variables were essentially only three (although they were duplicated) while the remaining four, which are more numerous, were given less significance.

65

To rectify this, it is necessary to give coefficients for each category of variables, according to their significance. The purpose of this is to distinguish the order of relevance of the variables and help make the analysis more logical.

66

7. Conclusion and Recommendations

7.1. Conclusion

Although FIs have introduced innovative services to help improve rural access to services, it is yet to be determined if these can contribute to increased access in the rural economy. Since the FIs' website does not contain other important information, such as number of clients, penetration rates, etc..., much attention has been given to agricultural loans, gender equality, and digital solutions in this study.

First, this investigation was based on a database of 144 FIs from the literature review, including 67 FIs from Africa, 29 from Asia, 35 from South America, and 13 IFIs. An analysis of the financial institutions reveals that, internationally, 58 percent offer credit to farmers, and 19 percent offer credit to value chain actors. Moreover, there are no restrictions on the availability of credit for small-scale farmers among the agricultural banks, international funds, and NGOs in this database, although local cooperatives and microfinance institutions are more likely to offer it. This confirm the Hypothesis H1 that Various trends can be seen based on the variables concerning the provision of credit by different types of FIs in the Global South.

Second, to enhance rural finance, there is a crucial need to ensure farmers have access to credit, promote the empowerment of women and promote the use of digital technology. It has been found that while 58 percent of FIs currently provide some type of agricultural credit, only 31 percent of FIs have programs geared toward empowering women, and 42 put digital solutions on the table. This statement answers the second research question and confirm the hypothesis H2, where not most FI deal with gender issues and offer digital solutions.

In these circumstances, the FIs are limited in their ability to request rural financial services. In addition, the results of the study are only able to provide some indication on the number, availability, and link between agricultural loans, gender, and digital solutions, but not on their impact, and therefore cannot accurately assess the potential of those services in improving access to agriculture.

Third, when examining the relationship between agricultural credit offers and the other variables, the differences and similarities between financial institutions could be determined through the method cluster analysis, which identified five clusters according to their main patterns. This cluster analysis classified all the observations into five groups, starting first with those who are value chain oriented. A second segment was corresponding with personnel with higher gender equality. The third segment mostly serves small farmers. Among the characteristics of the fourth cluster is its youthfulness and digital orientation. The fifth group

67

provides the minimum services due to its Traditional Banking approach. Analyzing the similarities and differences between these FIs provided an answer to the third research question as well as confirmed the hypothesis H3.

In the study, the most available information in a financial institution's website is geographic location, institutional type, year of foundation, number of branches, agricultural loans offered, gender equity and digital solutions. However, cluster analysis reveals that this relationship with digital solutions was not statistically significant, even when financial institutions were high in digital solution rates. In addition, it was revealed that there were no differences in agricultural finance offers depending on the number of branches or years of banking experience.

A critical feature of this database is that it offers a new outlook on rural finance services in the Global South. This is evidenced by rural financial institutions offering enhanced services designed in line with the needs of women. In addition to the finance gap in the agriculture sector, the limited adoption of digital products by FIs highlights a digital gap. Agricultural finance cannot be fully effective without more credit facilities for women, and innovative digital solutions.

7.2. Recommendations

The analysis of the database contributes to broadening the understanding of the potential of credit provision, gender, and digital solutions in rural finance. In this sense, the assessment allows offering some recommendations for further analysis:

As spoken by the professor Muhammed Yunus at the Australian Business Chambers Forum in Melbourne: ««While we celebrated, we noticed something. The money we gave to women was more effective. It did better to the family.»»

In this regard, as part of the rural finance, farmer credit should not be considered as the single solution to the challenges that small-scale farmers face. In fact, gender could play a role in improving the farmers' access to credit but need to be supported by other factors that are determinant for the operation of these models such as digital tools. There is a need for digital solutions to cater to the needs of small farmers. Some e-products (Email Banking, SMS Banking) will enable farmers to manage and consult their accounts, transfer money easily. In addition, online credit applications can reduce transaction costs greatly. A database of information about farmers should be collected by FIs to ease the lending process. Another effective tool for improving rural finance is the AVCF, which reduces transaction costs, manages risks, and offers better opportunities to value chain actors.

68

References

Abbas, A. C., Yuansheng, J., Feng, W. A., & Liu, D. (2017). Famers' access to credit: Does collateral. (A. Goodness, Ed.) Cogent Economics & Finance. doi: https://doi.org/10.1080/23322039.2017.1369383

Abid, A., Jie, S., Aslam, W., Batool, S., & Lili, Y. (2020). Application of structural equation modelling to develop a conceptual model for smallholder's credit access. The mediation of agility and innovativeness in organic food value chain finance. PLOS ONE, 15(8), e0235921. doi: https://doi.org/10.1371/journal.pone.0235921

Adams, D. W. (1994). Using contracts to analyse informal finance. (F. J. Bouman, & O. Hospes, Eds.) Financial landscapes reconstructed: The fine art of mapping development, 179-192.

Adebayo, O. O., & G, A. R. (2008). Sources and Uses of Agricultural Credit by Small Scale Farmers inSurulere Local Government Area of Oyo State. Anthropologist, 10(4), 3134.

AFD, INSE & IDFC. (2021, 07 20). Key Indicators by bank. Retrieved from Finance in Common: https://afdshiny.shinyapps.io/developmentbanksdatabase/

AfDB. (2013). Agricultural Value Chain Financing (AVCF) and Development for Enhanced

Export Competitiveness (Vols. (Vol. 136) [E-book]). African Development Bank.

AFRACA and FAO. (2020). Agricultural value chain finance innovations and lessons: Case studies in Africa. doi:10.4060/ca6345en

Akpan, S. B., Patrick, I. V., Udoka, S. J., Offiong, E. A., Okon, U. E., & Comment, G. (2013). Determinants of Credit Access and Demand among Poultry Farmers in Akwa Ibom State, Nigeria. American Journal of Experimental Agriculture,, 3(2), 293 307. doi:10.9734/AJEA/2013/2810

Ali, J., & Kumar, S. (2011). Information and communication technologies (ICTs) and farmers' decision-making across the agricultural supply chain. 31(2), 149-59.

Aliero, H. M., & Ibrahim, S. (2011). An analysis of farmers access to formal credit in the rural areas of Nigeria. African Journal of Agricultur al Research, 6(17).

Anderson, J. (2017). National Survey and Segmentation of Smallholder Households in Bangladesh: Household Level Data. Ref: BGD_2016_SHS_v01_M. Washington, D.C: World Bank. Retrieved 8 1, 2021, from https://microdata.worldbank.org/index.php/catalog/2839/study-description#metadata-identification

Antonaci, L., Demeke, M., & Vezzani, A. (2014). The challenges of managing agricultural

price and production risks in sub-Saharan Africa. ESA Working paper, 14-09. Retrieved from http://www.fao.org/3/i3907ee/i3907ee.pdf

Anwarul, H., & Prerna, T. (2015). Credit Policy for Agriculture in India - An Evaluation. (ICRIER, Ed.) Supporting Indian Farms the Smart Way: Rationalising Subsidies and Investments for Faster, Inclusive and Sustainable Growth.

Arias, J. (2019). Financing and financial inclusion for agricultural and rural transformation. IICA.

70

Bibliography

Asian Development Bank. (2012). Support for agricultural value chain development. (Manilla, Ed.)

Brown, M. Z. (2010). The emergence of information sharing in credit markets. 19(2), 255-78.

Buckley, G. (1997). Microfinance in Africa: Is it either a Problem or the Solution. World Development, 25(7), 1081-1093.

Cao, Z., & Lumineau, F. (2015). Revisiting the interplay between contractual and relational governance: A qualitative and meta-analytic investigation. 33, 15-42.

Carroll, T., Stern, A., Zook, D., Funes, R., Rastegar, A., & Lien, Y. (2012). Catalyzing

smallholder agricultural finance. Washington, DC: Dalberg Global Development Advisors.

CGAP. (2016). CGAP Smallholder Families Data Hub. Retrieved from Consultative Group to Assist the Poor: https://www.cgap.org/small_holders_data_portal/

Cheung, S. (1987). Economic organization and transaction costs.

Coleman, B. E. (1999). The Impact of Group Lending in Northeast Thailand. Journal of Development Economics, 60, 105-141.

Coon, J., Campion, A., & Wenner, M. (2010). Financing agriculture value chains in Central America (IDB). Washington, DC: Inter-American Development Bank.

Crook, J. N. (1996). Credit scoring: An overview. The University of Edinburgh: Department of Business Studies.

Cuevas, C., & Pagura. (2016). AGRICULTURAL VALUE CHAIN FINANCE A GUIDE FOR BANKERS. World Bank Group. Retrieved from https://www.mfw4a.org/sites/default/files/resources/Bankers_Guide_to_AVCF.pdf

D'Espallier, B. (2012). Unsubsidized Microfinanoe Institutions. Economics Letters, 120(2).

Da Silva, C. A., & Rankin, M. (2013). Contract farming for inclusive market access. Rome: Food and Agriculture Organization of the United Nations.

Dados, N., & Connell, R. (2012). The Global South. American sociological Association, 11(1),

12-13. doi:10.1177/1536504212436479

Das, A., Senapati, M., & John, J. (2009). Impact of agricultural credit on agriculture production: An empirical analysis in India. Reserve Bank of India Occasional Papers, 2(30), 75-107.

Diagne, A., & Zeller, M. (2001). Access to Credit and its Impacts in Malawi. Washington D.C: International Food Policy Research Institute.

Dzadze, P., Osei, M. J., Aidoo, R., & Nurah, G. K. (2012). Factors determining access to formal credit in Ghana: A case study of smallholder farmers in the Abura Asebu Kwamankese district of central region of Ghana. Journal of Development and Agricultural Economics, 14, 416 423.

Ellis, F. (1992). Ellis F. Agricultural policies in developing countries. FAO and ALIDE. (1996). Collateral in rural loans. Rome.

Ferdinand, A., & Asmah, E. (2012, June). A Brief survey on the literature on microfinance and Agriculture. Ghana Social Science Journal, 9(1).

71

Bibliography

Fries, B., & Akin, B. (2004). Value chains and their significance for addressing the rural

finance challenge. MicroREPORT, Accelerated Microenterprise Advancement Project.

Gamage, P. (2013). Determinants of access to bank finance for small and medium-sized enterprises: The case of Shri Lanka. Corporate Ownership and Control, 10(3), 402- 409. doi: https://doi.org/10.22495/cocv10i3c3art6

Gilbert, C. L. (2010). How to understand high food prices. J. Agric. Econ, 61, 398-425.

Glassman, C. A., & Wilkins, M. H. (1997). Credit scoring: probabilities and pitfalls. J. Retail Banking Service,.

Gonzalez, A., & Rosenberg, R. (2006). The State of Microfinance - Outreach, Profitability, and Poverty. Paper Presented at the Meeting Access to Finance: Building Inclusive Financial System. Washington, D.C: The World Bank.

Gowhar, B. A., Ganie, A. H., & Padder, M. J. (2013). A study on institutional credit to agriculture sector in India. (B. U. Department of Economics, Ed.) International Journal of current research and academic review, 1(4), 72-80. Retrieved from www.ijcrar.com

Grandori, A. F. (2019). Contracting for the unknown and the logic of innovation. 16(2), 413- 26.

Harhoff, D., & Korting, T. (1998). Lending relationships in Germany- empirical evidence from survey data. J. Banking and Finance, 22, 1317-1353.

Herliana, S., Acip, S., Qorri, A., Qonita, H. A., & Nur, L. (2018). The Constraints of Agricultural Credit and Government Policy Strategy. MATEC Web of Conferences, 215. doi:10.1051/matecconf/201821502008

Herliana, S., Sutardi, A., Aina, Q., Himmatul, A. Q., & Lawiyah, N. (2018). The Constraints of Agricultural Credit and Government Policy Strategy. MATEC Web of Conferences, 215, 02008. doi:10.1051/matecconf/201821502008

Honohan, & Beck. (2007). Making Finance Work for Africa,. Washington, D.C, 2007.

Hult, G. H. (2004). Innovativeness: Its antecedents and impact on business performance. (Vol. 5). Ind Mark Manag.

IFAD and CPI. (2020). IFAD, CPI, Chiriac, D., Naran, B., & Falconer, A. (2020, November). Examining the Climate Finance Gap for Small-Scale Agriculture. Climate Policy Initiative. https://www.ifad.org/documents/38714170/42157470/climate-finance-gap_smallscale_agr.pdf/34b2e25b-. Climate Policy Initiative. Retrieved from IFAD, CPI, Chiriac, D., Naran, B., & Falconer, A. (2020, November). Examining the Climate Finance Gap for https://www.ifad.org/documents/38714170/42157470/climate-finance-gap_smallscale_agr.pdf/34b2e25b-

IFC and World Bank. (2011). Scaling Up Access to Finance for Agricultural SMEs Policy Review and Recommendations. 84(2011th ed). Retrieved from https://www.gpfi.org/sites/gpfi/files/documents/G20_Agrifinance_Report%20%28FINA L%20ONLINE%29.pdf

IFC and World Bank. (2012). Innovative Agricultural SME Finance Models. Global

Partnership for Financial Inclusion. (2012th ed., Vol. 116). International Finance Corporation & World Bank.

72

Bibliography

Ijioma, J. C., & Osondu, C. K. (2015). Agricultural Credit Sources and Determinants of Credit, Acquisition by Farmers in Idemili Local Government. Journal of Agricultural Science and Technology. doi:doi: 10.17265/2161-6264/2015.01.004

Imtiaz SY, K. M. (2015). Telecom sector of Pakistan: Potential, challenges and business opportunities. 32(2), 254-8.

ISF Advisors and Mastercard Foundation. (2019). Pathways to Prosperity. Rural and Agricultural Finance State of the Sector Report, 61. Retrieved from https://docplayer.net/209577056-Pathways-prosperity-rural-and-agricultural-finance-state-of-the-sector-report-november-2019.html

Jones, G. E. (2015). Reconciling information gaps in organic farm borrowers' dealings with farm lenders. Agric Financ Rev, 75(4), 469-83.

Kaboski, J. P., & Townsend, R. M. (2012). The Impact of Credit on Village Economies. American Economic Journal: Applied Economics, 4(2).

Kaufman, A., Wood, C., & Theyel, G. (2000). Collaboration and technology linkages: a strategic supplier typology. 21(6), 649-63.

Khan, N., Shafi, M. M., & Shah, M. (2011). Review of past literature on agriculture credit in rural area of Pakistan. Sarhad J. Agriculture, 27(1), 103-110.

Kim, M., & Chai, S. (2017). The impact of supplier innovativeness, information sharing and strategic sourcing on improving supply chain agility. Global supply chain perspective, 187, 42-52.

KIT and IIRR. (2010). Value chain finance: Beyond microfinance tor rural entrepreneurs. Royal Tropical Institute, Amsterdam; and International Institute of Rural Reconstruction, Nairobi. Royal Tropical Institute, Amsterdam; and International Institute of Rural Reconstruction, Nairobi.

Klonner, S., & Rai, A. S. (2005). Adverse selection in credit markets. Evidence from South Indian bidding. Retrieved from

http://eml.berkeley.edu/~webfac/emiguel/e271_s06/klonner.pdf

Langenbucher, A. (2005). Warehouse receipt financing and related collateralized lending

mechanisms. Background briefing material prepared for the Commodity Risk Management Group.

Limsombunchai, V., Gan, C., & Lee, M. (2005). An Analysis of Credit Scoring for Agricultural Loans in Thailand. American Journal of Applied Sciences, 2(8), 1198-1205.

Lu, F. (1999). Three Grain Surpluses-Evolution of China's Grain Price and Marketing Policies (1978-1999). Symposium on China's Agricultural Trade and Policy, 251-268.

MapChart. (2021, 08 08). The Advanced World map adds political and geographical features to the basic World map. Retrieved from World Map: Advanced: https://mapchart.net/world-advanced.html

Meyer, R. (2011). Subsidies as an instrument in agriculture finance: A review. The International Bank for Reconstruction and Development. The World Bank.

Meyer, R. L. (2002). Track Record of Financial Institutions in Assisting the Poor in Asia. Retrieved from http://hdl.handle.net/11540/4154. License: CC BY 3.0 IGO

73

Bibliography

Meyer, R. L. (2007). Analysing and financing value chains: Cutting edge developments in value chain analysis. Paper presented at the 3rd African Microfinance Conference: New Options for Rural and Urban Africa. Uganda.

Miguel, D., & Silvana, P. (2007, 08 30). Commercial Banks and Microfinance. Munich Personal RePEc Archive. Retrieved from https://mpra.ub.uni-muenchen.de/10229/

Miller, C. (2012). Agricultural value chain finance strategy and design. Technical note. Agric value Chain Financ Strateg Des Tech note.

Miller, C., & Jones, L. (2010). Agricultural value chain finance Tools and Lessons (Vols. 195 [E-book]). (P. A. Publishing, Ed.) United Kingdom: Food and Agriculture Organization

of the United Nations. Retrieved from http://www.fao.org/3/i0846e/i0846e.pdf

Morduch, J. (1999). The Role of Subsidies in Microfinance: Evidence from the Grameen Bank. Journal of Development Economics, 60(1).

Mosley, P., & Hulme, D. (1998). Micro-enterprise Finance: Is there a Trade-off. Recent Research on Microfinance: Implications for Policy.

Muiruri, E. (2007). Strategic partnership for fi nance. presentation at the AFRACA Agribanks Forum.

Nyoro, J. (2007). Financing agriculture: Historical perspective. presentation at the AFRACA Agribanks Forum.

Ololade, R. A., & Olagunju, F. I. (2013). Determinants of Access to Credit among Rural Farmers in Oyo. Global Journal of Science Frontier Research, 13(2), 8. Retrieved 07 21, 2021

Osuntogun, A. (1980). Some aspects of farm level credit use in Nigeria. World Agricultural Economics and Rural Abstracts.

Pagano, M., & Jappelli, T. (1993). Information sharing in credit markets. 48(5), 1693-718.

Pearce, D. (2003). Rural finance innovation case study. Buyer and supplier credit to farmers: Do donors have a role to play? (BASIS-CRSP) Conference on Paving the Way Forward for Rural Finance, (p. Paper presented at Broadening Access and Strengthening Input market Systems). Washington DC, USA.

Petersen, M. A., & Rajan, G. R. (1994). The benefit of lending relationships: Evidence from small business data. J. Finance, 49, 3-38.

Piñeiro, V. (2019). Financing and financial inclusion for agricultural and rural transformation. (MTID, Ed.) Research Gate, 3. Retrieved from https://www.researchgate.net/publication/338804997

Plata, V., & Nartea, G. N. (1998). Credit analysis procedures of rural lenders in Canterbury. Paper. New Zealand: Agril. and Resource Econ. Soc.

Pradhan, N. C. (2013). Persistence of Informal Credit in Rural India: Evidence from All India Debt and Investment Survey and Beyond. Working paper No. 05.

Robert, F., Chalmers, G., & Grover, D. (2012). Value Chain Finance Guide: . Tools for Designing Project Interventions that Facilitate Investment in Key Value Chain UpgradesACDI/VOCA,. Retrieved from http://www.fao.org/sustainable-food-value-chains/library/details/en/c/267130/

74

Bibliography

Röttger, D. (2015). Agricultural finance for smallholder farmers: Rethinking traditional microfinance risk and cost management approaches.

Salami, A., & Arawomo, D. F. (2013). Empirical Analysis of Agricultural Credit in Africa: Any Role for institutional Factors? Working Paper no.192,.

Saqib, S. E., Kuwornu, K.M., J., Panezia, S., & Ubaid, A. (2018). Factors determining subsistence farmers' access to agricultural credit in flood-prone areas of Pakistan. Kasetsart Journal of Social Sciences, 39, 262-268.

Simon, A., & Yaya, L. (2012). Improving innovation and customer satisfaction through systems integration.

Singh, R. P. (2001). Rural credit and participation of credit agencies An analysis. Indian Cooperative Review, 38(4), 211-220.

Song, H. Y. (2018). Financial service providers and banks' role in helping SMEs to access finance.

Sriram, M. S. (2007). Productivity of Rural Credit. International Journal of Rural Management, 3(2), 245-268. doi:10.1177/097300520800300204

Svensson, G. (2001). Extending trust and mutual trust in business relationships towards a synchronised trust chain in marketing channels. Manag Decis.

Swafford, P., Ghosh, S., & Murthy, N. (2006). A framework for assessing value chain agility. Int J Oper Prod. Int J Oper Prod Manag.

Thai-Ha, L. (2020). Microfinance and Social Development: A Selective Literature Review. Fulbright University Vietnam, Fulbright School of Public Policy and Management. Asian Development Bank (ADB).

Turvey, C., & Brown, R. (1999). Credit scores for a federal lending institution: The case of Canada's farm credit corporation. Agril. Finance(50), 47-57.

UNCTAD. (2004, Novermber 16-17). Financing commodity based trade and development. Innovative agriculture financing mechanisms. Retrieved from tinyurl.com/y8azepg

Vandeginste, B., Massart, D., Buydens, L., De Jong, S., Lewi, P., & Smeyers-Verbeke, J.

(1998). Cluster analysis. Data Handling in Science and Technology, 57-68. doi:10.1016/s0922-3487(98)80040-3

Villalba, R., Venus, T., & Sauer, J. (2021). Agricultural Value Chain Financing: building a framework for an Ecosystem Approach. AARES Annual Conference. Australian Agricultural and Resource economics society.

WBG & WFDFI. (2018). 2017 Survey of national development banks. Malaysia: world bank group.

Wenner, M. (2006). Lecciones aprendidas en el fi nanciamiento de las cadenas agrícolas de valor: El caso del BID. presentation at the Latin American.

WOCCU. (2009). Technical Guide Integrated Financing for Value Chains. Washington, DC: World Council of Credit Unions.

WOCCU. (2021). Direct Members. Retrieved from World Council of Credit Unions: https://www.woccu.org/our_network/membership

75

Bibliography

World Bank. (1975). Agriculture Credit Sector Policy Review. (W. Bank, Ed.)

World Bank. (2007). Trade, Production, and Protection Database 1976-2004. The World Bank Economic Review.

World Bank and CGAP. (2017, December 13). Smallholder Families Data Hub. (J. Anderson, Editor, & Consultative Group to Assist the Poor) Retrieved July 31, 2021, from CGAP: https://www.cgap.org/small_holders_data_portal/

Xie, H. L., He, Y. F., & Xie, X. (2017). Exploring the factors influencing ecological land change for China's. J. Clean. Prod., 142, 677-687.

Xie, H. L., He, Y. F., Zou, J. L., & Wu, Q. (2016). Spatio-temporal difference analysis of cultivated land use intensity based on emergy in the Poyang Lake Eco-economic Zone of China. J. Geogr. Sci., 26, 1412-1430.

Yadav, P., & Sharma, A. (2015). Literature, Agriculture Credit in Developing Economies: A Review of Relevant. International Journal of Economics and Finance, 7(12), from 219. doi:10.5539/ijef.v7n12p219

Yuan, Y., Hu, Y., & Gao, P. (2011, July 24-26). Farmers' choice and informal credit markets in China. Paper presented at the Agricultural & Ap plied Economics Association s

2011 AAEA & NAREA Joint Annual Meeting, Pittsburgh,Pennsylvania, July 24.

Yunus, M. (2007). Creating a world without poverty. Social business and the future of capitalism. New York: Public Affairs.

Yunus, M., & Jolis, A. (1999). Banker to the Poor: Micro-Lending and the Battle against World. Public Affairs.

Zander, R. (1994). Barriers of credit access in rural Sri Lanka. (F. J. Bouman, & O. Hospes, Eds.) Financial landscapes reconstructed: The Fine Art of Mapping Development, 139-208.

Zander, R. (2016). Risks and Opportunities of Non-Bank-Based Financing for Agriculture: the Case of Agricultural Value Chain Financing. Deutsches Institut für Entwicklungspolitik GmbH. Bonn: Deutsches Institut für Entwicklungspolitik and German Development Institute. Retrieved 07 18, 2021, from https://www.die-gdi.de/uploads/media/DP_7.2016.pdf

Zhao, Q., Tsai, P.-H., & Wang, J.-L. (2019). Improving financial service innovation strategies for enhancing china's banking industry competitive advantage during the fintech revolution:. A Hybrid MCDM model, 11(5), 14-19.

Zhou, K., & Poppo, L. (2010). Exchange hazards, relational reliability, and contracts in China: The contingent role of legal enforceability. 41(5), 861-81.

76

Bibliography

Annex A

CEO E-mail-form

English	Dear Sir/Madam CEO I am Mohamed Ali Trabelsi, a master student at the Technical University of Munich (TUM School of Life Sciences). I am doing my master thesis under
	the supervision of Prof. Dr. Johannes Sauer, Roberto Villalba, and Terese
	Venus.
	This thesis consists of an online survey with financial institution on agricultural finance and the implication of the "agriculture value chain finance" approach. I am looking forward for your support for the realization of the survey, and I would be grateful if you can answer this online survey. You will find the link here and note that all information remains confidential.
	Thank you very much in advance. Sincerely, Mohamed Ali Trabelsi
French	Cher Monsieur le directeur général, chère Madame la directrice générale, Je suis Mohamed Ali Trabelsi, étudiant en master a l'université technique de Munich (TUM School of Life Sciences). Je suis entrain de faire ma
	mémoire de master sous la supervision de Monsieur Prof. Dr. Johannes
	Sauer, Monsieur Roberto Villalba, et madame Terese Venus.
	Cette mémoire consiste en une enquête en ligne avec les instituts financière sur le financement agricole et l'implication de l'approche "agriculture value chain finance". Je vous sollicite votre support pour la réalisation de l'enquête et je vous demande votre accord pour répondre a cette enquête en ligne. Le lien vous trouvez ici, et notez que toutes les informations restent confidentielles. Je vous remercie infiniment d'avance. Cordialement, Mohamed Ali Trabelsi

I

E-mail Form

Annex B: Database

II

Continent

	Financial Institution	Country	Type	Foundation	Branches
Africa	Cameroon Cooperative Credit Union League	Cameroon	cooperative	1968	12
Africa	Société Générale Côte d'Ivoire	Côte d'ivoire	Commercial Bank	1962	28
Africa	Banque Nationale D'Investissement	Côte d'ivoire	Commercial Bank	1959	22
Africa	Ecobank	Côte d'ivoire	Commercial Bank	1989	47
Africa	Société Ivoirienne de Banque	Côte d'ivoire	Commercial Bank	1962	66
Africa	Nouvelle Société Interafricaine d'Assurance	Côte d'ivoire	Insurance company	1906	0
Africa	Advans Côte d'Ivoire	Côte d'ivoire	MFI	2005	9
Africa	MicroCred Côte d'Ivoire	Côte d'ivoire	MFI	2010	23
Africa	MUCREFAB - Microfinance	Côte d'ivoire	MFI	1994	9
Africa	Fonds International pour le Développement de la Retraite Active	Côte d'ivoire	MFI	2011	8
Africa	Union Nationale des Coopératives d'Epargne et de Crédit de Côte d'Ivoire	Côte d'ivoire	MFI	1976	4
Africa	Eswatini Association of Savings and Credit Co-operatives	Eswatini	cooperative	1964	40
Africa	Nib International Bank	Ethiopia	Commercial Bank	1999	189
Africa	Harbu Microfinance Institution Share Company	Ethiopia	MFI	2005	13
Africa	Commercial Bank of Ethiopia	Ethiopia	Islamic Bank	1942	1714
Africa	National Association of Cooperative Credit Unions of The Gambia	Gambia	cooperative	1992	6
Africa	Ghana Co- operative Credit Union Association Ltd.	Ghana	cooperative	1955	11
Africa	Equity Bank	Kenya	Commercial Bank	1984	10
Africa	Agricultural Finance Corporation	Kenya	Agricultural Bank	1963	8
Africa	Faulu Kenya	Kenya	MFI	1991	67

Database

Africa

Family Bank

Kenya

Commercial Bank

1984

92

Africa

Cooperative Bank

Kenya

Commercial Bank

1931

100

Africa

Unity Finance

Kenya

cooperative

1974

15

Africa

Eclof Kneya

Kenya

MFI

1994

43

Africa

Juhudi Kilimo

Kenya

MFI

2004

39

Africa

Malawi Union of Savings and Credit Co-operatives, Ltd.

Malawi

cooperative

1962

47

Africa

Millennium bim - Particulares

Mozambique

Commercial Bank

1995

138

Africa

socremo banco de microfinanças

Mozambique

Commercial Bank

1998

13

Africa

African Banking Corporation

Mozambique

Commercial Bank

1999

10

Africa

Banco Mais

Mozambique

Commercial Bank

1999

7

Africa

Banco Único

Mozambique

Commercial Bank

2010

2

Africa

Barclays Bank

Mozambique

Commercial Bank

1991

3

Africa

Central Bank of West African States

Niger

International FI

1962

3

Africa

SunTrust Bank

Nigeria

Commercial Bank

2009

8

Africa

Keystonebank

Nigeria

Commercial Bank

1981

154

Africa

Union Bank of Nigeria

Nigeria

Commercial Bank

1917

320

Africa

UBA Group Corporate

Nigeria

Commercial Bank

1948

67

Africa

Zenith Bank

Nigeria

Commercial Bank

1990

390

Africa

Guaranty Trust Bank

Nigeria

Commercial Bank

1990

231

Africa

Standard Chartered Nigeria

Nigeria

Commercial Bank

1999

42

Africa

Polaris Bank Limited

Nigeria

Commercial Bank

2018

260

Africa

Heritage Bank Plc

Nigeria

Commercial Bank

1875

50

Africa

Bank of Kigali

Rwanda

Commercial Bank

1966

79

Africa

Banque Populaire Du Rwanda

Rwanda

Commercial Bank

1975

193

Africa

Association of Microfinance Institutions in Rwanda

Rwanda

MFI

2007

185

Africa

Seychelles Credit Union Asia

Seychelles

Credit union

1970

3

Africa

Development Bank of Southern Africa

South Africa

Development Bank

1983

0

Africa

Land and Agricultural

South Africa

Development Bank

1912

9

Database

III

Development Bank of South Africa

Africa

Equity Bank South Sudan Limited

South Sudan

Commercial Bank

2009

5

Africa

National Microfinance Bank

Tanzania

Commercial Bank

1997

226

Africa

FBME LIMITED

Tanzania

Commercial Bank

1990

4

Africa

Amana Bank

Tanzania

Islamic Bank

2011

10

Africa

Savings and Credit Cooperative Union League of Tanzania Ltd.

Tanzania

cooperative

1992

0

Africa

Azania Bank Ltd

Tanzania

Commercial Bank

1995

27

Africa

CRDB Bank

Tanzania

Commercial Bank

1996

260

Africa

Centenary Bank

Uganda

MFI

1985

63

Africa

Stanbic Bank

Uganda

Commercial Bank

1991

15

Africa

Mercantile Credit Bank

Uganda

Commercial Bank

1981

0

Africa

Uganda Cooperative Savings & Credit Union, Ltd.

Uganda

cooperative

1972

100

Africa

FINCA

Uganda

MFI

1992

29

Africa

UGAFODE Microfinance Limited (MDI)

Uganda

MFI

1994

19

Africa

First National Bank

Zambia

Commercial Bank

1838

22

Africa

ZANACO Bank

Zambia

Commercial Bank

1969

60

Africa

Stanbic Bank

Zambia

Commercial Bank

1906

63

Africa

Barclays Bank

Zambia

Commercial Bank

1991

47

Africa

National Association of Savings and Credit Unions

Zambia

cooperative

1969

0

Africa

National Association of Cooperative Savings & Credit Unions of Zimbabwe

Zimbabwe

cooperative

1980

118

Asia

Islamic Investment and Finance Cooperatives Group

Afghanistan

cooperative

2009

18

Asia

BRAC Bank

Bangladesh

Commercial Bank

2001

9

Asia

Bangladesh Commerce Bank Ltd

Bangladesh

Commercial Bank

1998

48

Database

IV

Asia

Prime Bank Limited

Bangladesh

Commercial Bank

1995

146

Asia

ONE Bank Ltd.

Bangladesh

Islamic Bank

1999

103

Asia

Standard Bank Ltd

Bangladesh

Islamic Bank

1999

34

Asia

sonali bank

Bangladesh

Islamic Bank

2007

1228

Asia

Grameen Bank

Bangladesh

MFI

1983

2568

Asia

HDFC Bank

India

Commercial Bank

1994

5653

Asia

ational Bank for Agriculture and Rural Development

India

Development Bank

1982

32

Asia

National Dairy Development Board of India

India

cooperative

1965

2

Asia

Basix India

India

MFI

1996

50

Asia

KBS Bank

India

Commercial Bank

1996

29

Asia

YES Bank

India

Commercial Bank

2003

1000

Asia

Reserve Bank of India

India

State development bank

1935

4

Asia

South Indian Federation of Fishermen Societies

India

cooperative

1980

0

Asia

Small Industries Development Bank of India

India

Commercial Bank

1990

33

Asia

National Agricultural Cooperative Federation

Korea

cooperative

1974

86

Asia

Mongolian Confederation of Credit Unions

Mongolia

cooperative

2008

22

Asia

Nepal Federation of Savings & Credit Cooperative Unions. Ltd.

Nepal

cooperative

1963

34

Asia

National Cooperative Bank Ltd. (NCBL)

Nepal

cooperative

2003

373

Asia

Quedan and Rural Credit Guarantee Corporation

Philippines

cooperative

1978

14

Asia

National Confederation of Cooperatives

Philippines

cooperative

1977

119

Asia

Development Bank of the Philippine

Philippines

state development bank

1947

127

Asia

Philippine Federation of Credit Cooperative

Philippines

cooperative

1960

0

Asia

Federation of Thrift & Credit

Sri Lanka

cooperative

1978

985

Database

V

Cooperative Societies Ltd.

Asia

bank for agriculture and agricultural cooperatives

Thailand

Development Bank

1966

1327

Asia

Small and Medium Enterprise Development Bank

Thailand

Development Bank

2005

27

Asia

Cooperative Bank of Vietnam

Vietnam

cooperative

2013

100

America

Barbados Co- operative & Credit Union League Ltd.

Barbados

cooperative

1957

29

America

Belize Credit Union League Ltd.

Belize

cooperative

1956

7

America

Bermuda Credit Union Co-op Society

Bermuda

cooperative

1972

0

America

Sartawi Foundation

Bolivia

MFI

1989

40

America

PROFIN Foundation

Bolivia

MFI

2001

0

America

Pro-rural

Bolivia

cooperative

2001

0

America

Sicredi Participações

Brazil

cooperative

2000

2000

America

Cayman Islands Civil Service Association Cooperative Credit Union Ltd.

Cayman Islands

cooperative

1975

46

America

Federación Nacional de Cooperativas de Ahorro y Crédito Financieras

Colombia

cooperative

2012

3205

America

Banco BAC San José S.A.

Costa Rica

Commercial Bank

1952

11

America

Federación de Cooperativas de Ahorro y Credito de Costa Rica R.L.

Costa Rica

cooperative

1999

14

America

The Curacao Federation of Cooperatives

Curacao

cooperative

1961

6

America

Asociación de Instituciones Rurales de Ahorro y Crédito, Inc.

Dominican Republic

cooperative

1991

24

America

Federación de Asociaciones Cooperativas

El Salvador

cooperative

1966

130

America

Cooperativas Micoope

Guatemala

cooperative

1963

300

America

Guyana Co- operative Credit Union League

Guyana

cooperative

2009

29

America

National Bank for Agricultural Development

Honduras

Agricultural Bank

1950

10

Database

VI

America

fundacioncovelo

Honduras

MFI

1985

0

America

VisionFund

Honduras

MFI

1995

17

America

Jamaica Co- operative Credit Union League Ltd

Jamaica

cooperative

1942

102

America

Banco Mercantil del Norte

Mexico

Commercial Bank

1899

20

America

Trust Funds for Rural Development

Mexico

Development Bank

1954

131

America

Nacional Financiera

Mexico

Development Bank

1934

3

America

Caja Popular Mexicana

Mexico

cooperative

1951

20

America

St. Patrick's Co- operative Credit Union

Montserrat

cooperative

2009

0

America

Banco Lafise

Nicaragua

Commercial Bank

1985

47

America

Credinka

Peru

cooperative

1994

90

America

Financiera Confianza

Peru

MFI

1991

115

America

Banco de Credito del Peru

Peru

Commercial Bank

1889

370

America

Fovida

Peru

cooperative

1987

3

America

Banco Bilbao Vizcaya Argentaria/Peru

Peru

Commercial Bank

1951

300

America

St. Kitts & Nevis National Co- operative League Ltd.

Saint Kitts & Nevis

cooperative

2009

4

America

St. Lucia Co- operative League Ltd.

Saint Lucia

cooperative

1977

15

America

St. Vincent and the Grenadines Co- operative League, Ltd.

Saint Vincent & the Grenadines

cooperative

1962

4

America

Co-operative Credit Union League of Trinidad & Tobago

Trinidad & Tobago

cooperative

1947

80

Global

Root Capital

International

NGO

1999

4

Global

Inter-American Development Bank

International

Development Bank

1959

28

Global

Rabobank

International

cooperative

1895

428

Global

International Finance Corporation

International

International FI

1956

184

Global

Incofin Investment Management

International

International Fund

2001

5

Global

Grameen Foundation and Fair Trade international

International

NGO

1997

2568

Global

Shared Interest

International

cooperative

1990

51

Database

VII

Global

Financial services and enterprise development

International

International Fund

1968

13

Global

European Microfinance Platform

International

MFI

1900

130

Global

Triodos Bank

International

Development Bank

1980

109

Global

International Raiffeisen Union

International

cooperative

1968

52

Global

International Fund for Agricultural Development

International

International FI

1977

177

Global

Inter-American Foundation

International

International Fund

1969

22

Database

VIII

Annex C: Survey for FIs Officials

Dear colleague,

We at the Technical University of Munich (TUM), Germany are conducting a survey about financial institutions. We would like to understand how your institution finances the agriculture sector, specifically financial products, the credit screening and scoring mechanisms employed, and the instrument used to mitigate risk and reduce transaction costs.

ü Why have I been chosen? Following a bibliographic search, we have identified your institute as meeting certain criteria eligible for this study

ü What will I be asked? general questions about your institute, the economic situation, and the method of financing the agricultural sector.

ü How my data will be stored? All the information we obtain will remain strictly confidential and anonymous.

Researcher's details: Mohamed Ali Trabelsi, Email: [e-mail] More Information: Supervisors: Roberto Villalba and Terese Venus.

If you would like further details, please check the university [TUM School of Life Sciences] website or this YouTube Link.

By consenting to take part, you agree to the following.

I have read the information sheet.

I have had the opportunity to ask question I understand my data will be anonymous

Q0. The survey will take about 20-30 minutes. Do you consent to take part? Choose an item. FORCE RESPONSE

ASK ONLY IF YES IS SELECTED IN Q0

1.0 General Information

Q1.1 How would you best describe your institution? Choose an item.

Q1.2 Please select the Continent: Choose an item.

Q1.3 Please select the Country Choose an item.

Q1.4 When was your institutions established Choose an item.

Q1.5 Could you select your position please? Choose an item.

Q1.6 How many years have you been working in agricultural finance? Choose an item.

Q1.7 How many years has your institution been working in agricultural finance? Choose an item.

Survey with Financial Institutions in the Global South - Multiple-respondent questionnaire	IX

2.0 Economic information

Q2.1 What where the total assets of your institutions? Click or tap here to enter text. Q2.2 What was your institution's total equity? Click or tap here to enter text.

Q2.3. What was the total gross loan portfolio of your institution? Click or tap here to enter

text.

Q2.4 How many branches did your institution have? Click or tap here to enter text. Q2.5 How many employees did your institution have? Click or tap here to enter text. Q2.6 Which one of the following is the priority sector Choose an item.

3.0 Agricultural credit: screening, scoring and monitoring

Q3.1 Do you offer credit to all your customer? Choose an item. FORCE RESPONSE ASK ONLY IF YES IS SELECTED IN Q3.1

*Product and Financial flows*

Q3.2 What lending products does your institution offer?Check all that apply.

ALLOW MULTIPLE ANSWERS

3.2.1. Agricultural loan	?
3.2.2. Animal Husbandry loan	?
3.2.3. Consumption loan	?
3.2.4. Production loan (input)	?
3.2.5. Income generating activities	?
3.2.6. Working capital loans	?
3.2.7. Asset building	?
3.2.8. Lease financing	?
3.2.9 Micro-leasing (equipment)	?
3.2.10. Emergencies	?
3.2.11. Commercialization credit	?
3.2.12. Factoring (delay payment)	?
3.2.13. Others	?

Q3.3 What is the average loan per Borrower or range does your institution offer to client? Choose an item.

Q3.4 What time frame for loans does your institution provide? ALLOW MULTIPLE ANSWERS

3.4.1. Short-term Loan	?
3.4.2. Medium-term Loan	?
3.4.3. Long-term Loan	?

X

Survey with Financial Institutions in the Global South - Multiple-respondent questionnaire

Q3.5 How is the loan disbursed? Choose an item.

Q3.6 How must the loan be repaid? Choose an item.

Q3.7 What is the interest rate per month?Choose an item.

Q3.8 What is the interest rate per year?Choose an item.

Q3.9 What is the percentage of other additional if incurred Choose an item.

*Risk management* Q3.10 How is the loan securitized? ALLOW MULTIPLE ANSWERS
3.10.1 Securitization based on sales contract	?
3.10.2 Loan is percentage of value of produce	?
3.10.3 Collateral provided	?
3.10.4 Transaction carried out through bank account	?
3.10.5 Warehouse inventory	?
3.10.6 Deposit	?
3.10.7 Producer organizations members (group account)	?
3.10.8 Assessment and investigation	?
3.10.9 Legal document	?
3.10.10 Household Item/ Leased Item	?
3.10.11 Peer pressure from group, pledged to group	?
3.10.12 Borrower must own the land	?

Q3.11 Who has to repay the loan after lending?

3.11.1. Individual Farmer/Borrower	?
3.11.2. Cooperative	?
3.11.3. Farmer group	?
3.11.4. neighborhood groups	?
3.11.5. Association	?
3.11.6. Marketing organization	?
*Information flows* Q3.12 Which of the following information does the institution require before the loan is disbursed? ALLOW MULTIPLE ANSWERS
3.12.1 recent audited financial statement			?
3.12.2. recent interim financial statement			?
3.12.3. Cash flow prognosis for new season			?
3.12.4. Projected profit			?
3.12.5. Review of previous season (yield, contract, etc.)			?
3.12.6. development of market prices			?
3.12.7. realized investment			?
3.12.8. identity card			?
3.12.9. status in registry financial systems			?
3.12.10. Registration of membership			?
Survey with Financial Institutions in the Global South - Multiple-respondent questionnaire				XI
3.12.11. Basic plan on how the loan will be used		?
3.12.12. Credit history		?
3.12.13. Potential buyers		?
3.12.14. Group registration certificate		?
3.12.15. Application letter		?
3.12.16. current year management account		?
3.12.17. Account number		?
3.12.18. Reputation		?
3.12.19. Characteristics of the family		?
3.12.20. Number of Ha owned		?
3.12.21. Area planted		?

Q3.13 What additional information do you require during the loan period? ALLOW MULTIPLE ANSWERS

3.13.1. Monthly savings of group	?
3.13.2. Report of monthly meetings	?
3.13.3. A field credit officer	?
3.13.4. Check crop progress	?
3.13.5. Price of agricultural inputs	?
3.13.6. Farm inspection by specialist	?
3.13.7. Monthly project monitoring report	?

Q3.14 How long take on average between the application and the disbursement of the loan? Choose an item.

*Borrower characteristics* Q3.15 Which of the following affect the lending decision? ALLOW MULTIPLE ANSWERS
3.15.1. Assets	?
3.15.2. Age	?
3.15.3. Education	?
3.15.4. Marital status	?
3.15.5. Number of children	?
3.15.6. Employment	?
3.15.7. Farm type	?
3.15.8. Landholding size	?
3.15.9. Borrower's Liquidity	?
3.15.10. Repayment capacity	?
3.15.11. Relationship indicators	?
3.15.12. Collateral requirements	?
3.15.13. Province	?
3.15.14. Other features	?

Q3.16 What is the most suitable age for getting the credit? ASK ONLY IF AGE IS SELECTED IN 3.15

3.16.1. < 25-year-old ? 3.16.2. 26 - 35-year-old ? Survey with Financial Institutions in the Global South - Multiple-respondent questionnaire	XII

3.16.3. 36 - 45-year-old ?

3.16.4. >46-years-old ?

03.17 Which of the following education level is most reliable for credit allocation? ASK ONLY IF EDUCATION IS SELECTED IN 3.15

3.17.1. Illiterate	?
3.17.2. Literate	?
3.17.3. Primary	?
3.17.4. Middle Class	?
3.17.5. High School	?
3.17.6. Intermediate	?
3.17.7. Graduation	?
3.17.8. Post-Graduation and Above	?
3.17.9. Any professional course & Technical	?

03.18 What farmer's Marital Status do the financial Institution consider as an advantage before lending? ASK ONLY IF MARITAL STATUS IS SELECTED IN 3.15

03.18.1. Married	?
03.18.2. Unmarried	?
03.18.3. Widow	?
03.18.4. Divorced	?
03.18.5. Separated	?

03.19. What number of children is an advantage for an agricultural credit? ASK ONLY IF NUMBER OF CHILDREN IS SELECTED IN 3.15

3.19.1. None	?
3.19.2. 1-2 Children	?
3.19.3. 3-5 Children	?
3.19.4. > 6 children	?

03.20 Does your institute consider farmer Employment, if yes, which of the following is a key factor? ASK ONLY IF EMPLOYMENT IS SELECTED IN 3.15

3.20.1. Employed	?
3.20.2. Unemployed	?
3.20.3. self Employed	?
3.20.4. Professional	?

03.21 What types of agriculture activity (Farm type) eligible for a credit? ASK ONLY IF FARM TYPE IS SELECTED IN 3.15

3.21.1. Arable Farming	?
3.21.2. Pastoral Farming	?
3.21.3. Mixed Farming	?
3.21.4. Subsistence Farming	?
3.21.5. Commercial Farming	?

XIII

Survey with Financial Institutions in the Global South - Multiple-respondent questionnaire

3.21.6. Extensive and Intensive Farming ?

3.21.7. Nomadic Farming ?

3.21.8. Sedentary Farming ?

3.21.9. Poultry Farming ?

3.21.10. Fish Farming ?

Q3.22 How many Ha of agricultural land (Landholding Size) do farmers at least operate to be qualified for a loan? ASK ONLY IF LANDHOLDING SIZE IS SELECTED IN 3.15

3.22.1. Landless ?

3.22.2. < 1Ha ?

3.22.3. 1-5 Ha ?

3.22.4. 6-10 Ha ?

3.22.5. 11 Ha ?

Q3.23 What types of guarantees are accepted by your institute? ASK ONLY IF COLLATERAL IS SELECTED IN 3.15

3.23.1. Livestock ?

3.23.2. equipment ?

3.23.3. guarantor ?

3.23.4. Land ?

3.23.5. Others ?

Q3.24 Which of the relationship indicators does your institute take into consideration? ASK ONLY IF NUMBER OF RELATIOINSHIP INDICATORS IS SELECTED IN 3.15

3.24.1. Borrowing from others ?

3.24.2. Duration for repayment ?

3.24.3. Others ?

4.0 Agricultural finance within value chains

Q4.1 Do your institution use any of these practices?Choose an item.

Q4.2 Are you familiar with this approach of AVCF or have you ever heard about it? Choose an item. FORCE RESPONSE

ASK ONLY IF YES IS SELECTED IN Q4.3

5.0 Financial product & Instrument employed

Q5.1 Out of the following five financial products, which type of financial product are you familiar with/do your institution apply in agriculture lending? ALLOW MULTIPLE ANSWERS

5.1.1 Product financing ?

5.1.2 Receivables financing ?

5.1.3 Physical asset collateralization ?

5.1.4 Risk mitigation products ?

Survey with Financial Institutions in the Global South - Multiple-respondent questionnaire	XIV

5.1.5 Financial enhancements ?

ASK ONLY IF PRODUCT FINANCING IS SELECTED IN Q5.1

05.2 Which of the following instruments has <INSERT 01.1 INSTITUTION> ever used?

5.2.1 Trader credit	?
5.2.2 Input supplier credit	?
5.2.3 Marketing Credit Company	?
5.2.4 Lead firm financing	?

ASK ONLY IF RECCEIVABLES FINANCING IS SELECTED IN Q5.1

05.3 Which of the following instruments has <INSERT 01.1 INSTITUTION> ever used?

5.3.1 Trade receivables finance	?
5.3.2 Factoring	?
5.3.3 Forfaiting	?

ASK ONLY IF PHYSICAL ASSET COLLATERALIZATION IS SELECTED IN Q5.1

05.4 Which of the following instruments has <INSERT 01.1 INSTITUTION> ever used?

5.4.1 Warehouse receipts	?
5.4.2 Repurchase agreements	?
5.4.3 financial leasing	?

ASK ONLY IF RISK MITIGATION PRODUCTS IS SELECTED IN Q5.1

05.5 Which of the following instruments has <INSERT 01.1 INSTITUTION> ever used?

5.5.1 Insurance	?
5.5.2 Forward contracts	?
5.5.3 Futures	?

ASK ONLY IF FINANCIAL ENHANCEMENTS IS SELECTED IN Q5.1

05.6 Which of the following instruments has <INSERT 01.1 INSTITUTION> ever used?

5.6.1 Securitization instruments	?
5.6.2 Loan guarantees	?
5.6.3 Joint venture finance	?

047. Ask question/Feedback Click or tap here to enter text.

We thank you Sir/Madam <INSERT 01.6 Name> for your time spent taking the survey Your response has been recorded

*****

XV

Survey with Financial Institutions in the Global South - Multiple-respondent questionnaire

Annex D: R Script

# Descriptive Analysis #

data1 <- na.omit(X20210922_database_FIs_MAT)

data<- data1[-1]

plot(data)

data

summary(data)

#Scale Data #

Data_scaled = apply(data, 2, function(r) {

if (sd(r) != 0)

res = (r - mean(r))/sd(r) else res = 0 * r

res

})

summary(Data_scaled)

install packages#

install.packages("dplyr")

install.packages("ggplot2")

install.packages("ggfortify")

install.packages("factoextra")

install.packages("clusterSim")

# load required libraries #

library(stats)

library(dplyr)

library(ggplot2)

library(ggfortify)

library(factoextra)

# WSS plot function #

wssplot <- function(Data_scaled, nc=15, seed=1234){

R Script	XVI

wss <- (nrow(Data_scaled)-1)*sum(apply(Data_scaled,2,var))

for (i in 2:nc){

set.seed(seed)

wss[i] <- sum(kmeans(Data_scaled, centers=i)$withinss)}

plot(1:nc, wss, type="b", xlab="Number of Clusters",

ylab="Within groups sum of squares")}

# Hierarchical Clustering ....

d <- dist(Data_scaled)

fith <- hclust(d, "ward.D2")

plot(fith)

# WSS Plot to choose maximum number of clusters #

wssplot(Data_scaled)

# Choosing K

k<- list()

for (i in 1:10){

k[[i]]<- kmeans(na.omit(data), i)

}

k[[9]]

k

betweenss_totss <- list()

for (i in 1:10) {

betweenss_totss[[i]]<-k[[i]]$betweenss/k[[i]]$totss

}

plot (1:10, betweenss_totss, type = "b" ,

ylab = "between SS / Total SS", xlab = "clusters(k)")

for (i in 1:4){

plot(data, k[[i]]$cluster)

}
	R Script	XVII

# K-Means cluster #

KM= kmeans(Data_scaled,5)#

# Evaluating Cluster Analysis#

# Cluster plot 5#

autoplot(KM,Data_scaled,frame=TRUE)

# Segmentation & Visualization #

KM.clusters <- KM$cluster

rownames(Data_scaled) <- data1$id

fviz_cluster(list(data=Data_scaled,cluster=KM.clusters))

# cluster Membership #

KM$cluster

table(KM.clusters, data1$id)

table(data1$id, KM.clusters)

# Clusters Center #

KM$centers

# descriptive statistics at the cluster level

library(clusterSim)

desc <- cluster.Description(Data_scaled,KM$cluster)

print(desc)

summary(desc)

library(factoextra)

sil <- silhouette(KM$cluster, dist(Data_scaled))

fviz_silhouette(sil)

R Script	XVIII