9. :-
3.63..

3%.

 

 

 

 

. .. u.
u. ,Rtmwéafi}
0C. Lystnfiufin .

. 5.65%

a \ [flare
2%, r sway,

1
..

2T... ‘
fink“. w

3:. xx!!!
. liv‘li.‘
A :1"... 3

t It, .31:
. it... )!
7‘.ul=u\\.‘,’£

a v.63».

 

'21::
. w! v!
’v.‘ u‘ -.

7811.2...92 i
1.32.5.3... nus...
: . , xi...

p 5!?!3v

awaéafia finfi. afinmfiwngrw3fi . ,
‘ . V .IA .3.-

 

. 1 X
10.9 ; LSBRARY
; Articmgan State

University .

This is to certify that the
dissertation entitled

THE ROLE OF HOUSEHOLD PHYSICAL, HUMAN, NETWORK
AND ENVIRONMENTAL ASSETS IN SMALL FARMER
ACCESS TO MARKETS: Guatemalan Lettuce Growers’
Experiences with Modern Market Channels

presented by
LUlS G. FLORES

has been accepted towards fulfillment
of the requirements for the

Doctoral degree in Community, Agriculture,
Recreation and Resource
Studies

 

 

 

Major Pro‘t’essot’s Signature
5[ 15/ / 0

Date

 

MSU is an Afi‘innaiive Action/Equal Opportunity Employer

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:IProj/Aoc&Pres/C|RCIDateDue.indd

THE ROLE OF HOUSEHOLD PHYSICAL, HUMAN, NETWORK AND
ENVIRONMENTAL ASSETS IN SMALL FARMER ACCESS TO MARKETS:
Guatemalan Lettuce Growers’ Experiences with Modern Market Channels

By

Luis G. Flores

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Community, Agriculture, Recreation and Resource Studies

2010

ABSTRACT
THE ROLE OF HOUSEHOLD PHYSICAL, HUMAN, NETWORK AND
ENVIRONMENTAL ASSETS IN SMALL FARMER ACCESS TO MARKETS:
Guatemalan Lettuce Growers’ Experiences with Modern Market Channels

By

Luis G. Flores

There is no consensus on how increased globalized trade supports poverty
reduction in developing countries. Central to this developing debate is whether modern
market channels favor small farmer inclusion or exclusion from higher income
opportunities. In Guatemala, the increased importance of modem markets has been
studied with particular attention to multinational foodservice companies such as
McDonalds and supermarkets such as Wal-Mart. These new market channels are
characterized by high procurement standards in product quality, safety and delivery.
Unlike the experiences in other regions in the world, these high-standard supply chains
(HSSCS) have favored the participation of small—scale farmers who happen to be
producing fresh lettuce. The aim of this study is to develop a better understanding of the
evolution and operation of high—standard supply networks that favor small farmer
inclusion as well as develop an understanding the organizational and individual
characteristics of small farmers able to successfully participate in HSSCs. The study is
based on primary data collected in six villages of the Guatemala central highlands during
the summer and fall of 2008.

The study’s results reveal important insights concerning how successful small

farmers manage their organizations to compete and how they are able to emulate ‘

practices of privately-owned, larger companies. The estimated empirical model also
underlines the critical role of physical, human and network assets of the individual farm,
but also highlights the increasing importance of environmental characteristics of farmer

households for farmers’ access to more profitable market opportunities such as HSSCs.

DEDICATION

To my wife, Maripaz.

iv

ACKNOWLEDGEMENTS

I want to thank all members of my graduate committee for their time and dedication to
my dissertation project. I also want to express special thanks to my major professor, Dr.
Michael Kaplowitz, for believing in me. This work would have not been completed

without his technical support from the early drafting stages.

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... viii
LIST OF FIGURES ............................................................................................................. x
LIST OF ACRONYMS ...................................................................................................... xi
CHAPTER 1 ........................................................................................................................ 1
INTRODUCTION ............................................................................................................... 1
1.1 General objective of the dissertation ......................................................................... 1
1.2 Central-America’s participation in global trade ........................................................ 2
1.3 Major market development booms in Guatemala ...................................................... 4
1.4 Research objectives ................................................................................................... 6
References ....................................................................................................................... 9
CHAPTER 2 ...................................................................................................................... 1 1
A VALUE CHAIN ASSESSMENT OF GUATEMALA’S LETTUCE SUBSECTOR...11
2.1 Introduction ............................................................................................................. 11
2.1.1 Guatemalan small-holder agrarian systems ...................................................... 14
2.1.2 Guatemalan lettuce subsector ........................................................................... 15
2.2 Theoretical basis for value chain assessments ......................................................... 16
2.3 Value Chain Assessment Methods .......................................................................... 21
2.4 Findings ................................................................................................................... 23
2.4.1 The lettuce value chain networks ..................................................................... 23
2.4.2 Differences in farmers trust and coordination across market channels ............ 32
2.5 Conclusions ............................................................................................................. 38
References ..................................................................................................................... 50
CHAPTER 3 ...................................................................................................................... 55
A CASE STUDY OF SMALL FARMER ORGANIZATIONAL STRATEGIES TO
INCREASE MARKET ACCESS PERFORMANCE ....................................................... 55
3.1 Introduction ............................................................................................................. 55
3.2 Historical review of Labradores Mayas ................................................................... 57
3.3 Theoretical framework ............................................................................................ 60
3.4 Case study methods ................................................................................................. 65
3.5 Results ...................................................................................................................... 66
3.5.3 Access to HSSCs as an organizational shaping factor ..................................... 66
3.5.2 LM strategies to access HSSCs directly ........................................................... 68
3.6 Conclusions ............................................................................................................. 73
References ................................................................................................................. 79
CHAPTER 4 ...................................................................................................................... 82

SMALL FARMER’S HOUSEHOLD PHYSICAL, HUMAN, NETWORK AND
ENVIRONMENTAL ASSETS AS EXPLANATORY VARIABLES IN FARMERS
ACCESS TO HSSC’S ....................................................................................................... 82

vi

4.1 Introduction ............................................................................................................. 82

4.2 Prior market booms and the Guatemala lettuce subsector ....................................... 84
4.3. Literature review ..................................................................................................... 87
4.3.1 The role of assets in NTAX adoption—learning from previous research ........ 87
4.3.2 Modeling supermarket channel adoption after the NTAX experience ............. 91
4.3.3 Empirical evidence from Kenya guides analysis on HSSCs adoption ............. 91
4.4 Theoretical model .................................................................................................... 93
4.5 Empirical model specification ................................................................................. 94
4.6 Data and estimation ................................................................................................. 97
4.7 Results ................................................................................................................... 100
4.7.1 Descriptive statistics of physical, human, network and environmental assets
................................................................................................................................. 100
4.7.2 Econometric results and discussion ................................................................ 104
4.8 Conclusions ........................................................................................................... 109
References ................................................................................................................... 1 22
Appendix Chapter 4 ..................................................................................................... 125
CHAPTER 5 .................................................................................................................... 129
SUMMARY AND CONCLUSIONS .............................................................................. 129
APPENDIX ..................................................................................................................... 134
Value Chain Study Protocol ........................................................................................ 134
Open-ended Interview Guide ................................................................................... 136
Case Study Protocol ..................................................................................................... 138
Table A2 Summary of research issues and propositions ........................................ 138
Survey instrument ........................................................................................................ 14 1

vii

LIST OF TABLES

Table 2.1: 2007-2008 Market Share and Business Practices Gap Analysis ......... 40
Table 2.2: Historical Data on HSSCs ..................................................... 40
Table 2.3: Chronological development of HSSCs in Central America .............. 41
Table 2.4: Guatemalan Lettuce Suppliers Entry to GU and ES Markets ............ 42
Table 2.5: 2007-2008 Farm gate and Primary Gathering Site Cost Structure

US$/Kg) ....................................................................................... 42
Table 2.6: Perception of Small Farmer Assets GAP Analysis ........................ 43
Table 2.7: Guatemalan Lettuce“ Production and Value 1989-2006 (000's MT and

000's 8) ........................................................................................ 44
Table 3.1: Guatemalan Lettuce Suppliers Entry to GU and ES Markets ............. 74
Table 3.2: Perception of Small Farmer Assets GAP Analysis ......................... 75
Table 4.1: Guatemalan Lettuce* Production and Value 1989-2006 (000's MT and

000's $) ........................................................................................ 11 1
Table 4.2 Comparison of household physical asset characteristics in 2008 ......... 1 12
Table 4.3 Comparison of household physical asset characteristics in 2004 ......... 1 12
Table 4.4 Comparison of household physical asset characteristics 2001 ............ 113
Table 4.5 Comparison of household human assets within groups for 2008 ......... 114
Table 4.6 Comparison of household network asset between groups for 2008 ...... 1 15

Table 4.7 Comparison of household environmental asset between groups for

2008 ........................................................................................... 1 16
Table 4.8 Probit results on the effects of small farmer assets in accessing HSSCs

markets ........................................................................................ 1 17
Table 4.9 Hausman endogeneity test on total lettuce grown ........................... 124
Table 4.10 Hausman endogeneity test on membership in association(s) ............. 125
Table 4.11 Hausman endogeneity test on experience growing lettuce ............... 126

viii

Table 4.12 Summary of significance between physical assets among HSSCs and

Traditional Suppliers during 2008, 2004 and 2001 ................................... 127
Table A. 1. Research topics and expected number of interviewees and

determination of persons to be interviewed ............................................. 135
Table A2 Summary of research issues and propositions .............................. 138

ix

LIST OF FIGURES

Figure 2.1: The 2007-2008 Value Chain Map ............................................. 53
Figure 2.2: Wal-Mart Central America Store Formats and Number of Outlets ........ 54
Figure 2.3: Lettuce Tangible and Intangible Product Traits ............................ 55
Figure 2.4: Michael Porter’s (1886) Value Chain Structure ............................. 55
Figure 2.5: Research Site ...................................................................... 56
Figure 3.1: Lettuce Tangible and Intangible Product Traits ............................. 84
Figure 3.2: Small Farmer Organizations Roadmap to HSSCs ............................. 85
Figure 4.1: The 2007-2008 Lettuce Value Chain Map ...................................... 129
Figure 4.2: Lettuce Tangible and Intangible Product Traits ................................ 130
Figure 4.3: Research Site ....................................................................... 131

AGEXPORT

AT

APBF

CACM

C.A.

CARHCO

CENMA

CBI

ES

GDP

GT

HACCP

HSSCS

IMF

LM

NTAX

OECD

TCE

INE

MAGA

Mkt.

MT

LIST OF ACRONYMS

Guatemalan Association of Exporters
Agency Theory

Associative Peasant Business Firms
Central-America Common Market
Central-America

Central American Retail Holding Company
Central de Mayoreo (English: Central Produce Terminal)
Caribbean Basin Initiative

El Salvador

Gross Domestic Product

Guatemala

Hazard Analysis and Critical Control Points
High-Standards Supply Chains
International Monetary Fund

Labradores Mayas

Non-Traditional Export Crops

Organization of Economic Cooperation and Development
Transaction Costs Economics

[Guatemala] National Institute of Statistics
[Guatemala] Ministry of Agriculture
Market

Metric Ton(s)

xi

NGO Non-Governmental Organization
SIECA Sistema de Informacion Economica de Centre-America (English: Central
American Economic Information System)

Trad. Traditional

SWOT Strenghts-Weaknesses-Opportunites-Threats
UNDP United Nations Development Programme
WB World Bank

xii

CHAPTER 1

INTRODUCTION

1.1 General objective of the dissertation

One of the most important buzz words in the development literature of the last
three decades has been “globalization.” This term represents the expansion of trade and
foreign investment from domestic markets to a worldwide environment. In agriculture,
globalization and internationalization activities have been remarkable since the 19703
dealing with tariffs, grades and standards and regional agreements to expand the
exchange of commodities and specialties (Reardon and Flores 2007). Globalization has
improved flow of capital, ideas, people and technology; however, the impact of this
phenomenon on poverty, particularly on small-scale agriculture inclusion, exclusion and
income inequality in developing countries remains unknown and subject to investigation
(Bardhan 2004; Maertens et al 2008). Therefore, the aim of this dissertation is to
contribute to the literature on trade and poverty reduction by examining the case of small
farmers from the Guatemalan highlands, a geographic area where three major market
development booms have taken place since the advent of increased globalized trade.

The case of lettuce farmers is studied as a success story in the context of the
regional market development for supermarkets and restaurant chains in Guatemala. These
two market channels are known for their higher standards in terms of tangible and
intangible product traits such as quality, safety and delivery. Tangible product traits are
established on the basis of appearance such as maturity, size, presence of defect, and
packaging. Intangible traits are factors verifiable through supply performance and regular

field inspection such as compliance with food safety assurance activities mainly focused

on proper use of pesticides, potable water quality and worker’s hygiene (Calvin 2002;
Jano and Mainville 2006). Evidence from Central America and other areas in the world
points out procurement systems of these dynamic, more demanding markets have strict
procurement standards that often grow less favorable for small farmer participation.

In order to understand how small farmers have entered these markets, this
dissertation contributes to the literature through the development of three separate, but
interconnected research pieces. First, a value chain assessment of the lettuce subsector
presents a thorough assessment of the evolution of different market channels during the
past twenty years; their actors, market shares in domestic and regional markets and the
enabling environment that supported its growth domestically and regionally are also
studied. Second, a case study of a successful market organization is developed, assessing
the group- and individual farmer-based decision making process in accessing high-
standards supply chains (HSSCs). HSSCs are given this definition to separate them from
market channels with less strict procurement and product standards such as traditional,
open-air markets. Third, this paper contributes to the literature on assessing the role of
key small farmer’s assets as determinants that favor their participation in the HSSCs
under study. Primary data from a random sample of HSSCs and traditional market
suppliers was generated through a field survery collecting information on key physical,

human, network and environmental assets of small-scale lettuce farmers.

1.2 Central-America’s participation in global trade

Most Latin-American countries have implemented far-reaching structural reforms
to favor free trade since the 19805 (Meller 2008). This major change in economic

development strategy, initially stimulated and promoted by the World Bank (WB) and the

International Monetary Fund (IMF), has favored some countries more than others (Meller
2008). In the case of Central-America, preliminary results by the early 19905 in terms of
overall economic growth and income distribution, the region’s diverse climatic
conditions and strategic geographic location with respect to developed-country markets
favored adoption of export crop production (von Braun et al., 1989; James et al., 2000).
Specifically, from 1983 to 1997 horticulture exports (excluding bananas) to the US
market grew from $47 to $456 million in 1999 (James et al., 2000). Fruits and vegetables
alone, grew 16% annually from 1983 to 1997 being Guatemala, Costa Rica and Honduras
the major regional participants in this commercial expansion (James et al., 2000).

Three parallel factors have supported this agricultural trade expansion process.
First, this period has coincided with the end of civil and political unrest in the Central-
American region, particularly in Guatemala and El Salvador. Second, economies have
experienced low, but stable growth, favoring the business environment for foreign trade
and investment in several industries and the development of domestic and regional
markets. And third, regional projects financed by foreign aid were strategically designed
and executed to bring the necessary know-how from production to markets to support the
development of export crops as a major income-generation activity for the rural economy
(Lamb 1991). The linkages associated to this development have given rise to a number of
related industries (linkages) providing supporting goods and services throughout the
Central-American region. As a result, Central American economies became not just
suppliers but also consumers of products and services leading to the consolidation of the
Central-America Common Market (CACM) (Outreville 1996). Recently, small farmer

participation in these important market developments has been at the center of the debate

on the effect of globalized trade and poverty reduction (Berdegué et al., 2005; Balsevich
2006; Reardon and Flores 2007; Hernandez et al., 2007). This question is also addressed
in other regions of the world where similar market booms have—and are—taking place
as a result of emerging economies taking advantage of market developments domestically

and internationally (Okello 2006; Maertens and Swinnen 2007).

1.3 Major market development booms in Guatemala

Guatemala has been an important research scenario to analyze the effects of three
major agricultural market development booms with different effects on small farmers.
First, the commodity development boom from the 19505 through the 19703, particularly
cotton, cattle and sugar cane, was characterized by the hi gh-scale initial capital
investment requirements particularly for production land, processing infrastructure and
machinery. The production of these commodities also required tens of thousands of
laborers, many of whom were brought from the central and western highlands as
temporary workers during the harvesting season. Given the required start-up capital and
working capital needed to succeed in these commodities, the participation of large
landholding agrarian systems in the Pacific Littoral and Atlantic Coast was favored
(Williams 1986).

A second remarkable boom caused mainly by major shifts in supply patterns in
markets potentially accessible to Central American producers was the non-traditional
export crops (N TAX). Being most NTAX participation by Central-American growers
characterized by high risk and relatively high capital investments, small farmers were not

particularly targeted as adopters (Lamb 1991). However, the labor-intensive crops and

diversity of climates required by some of those products such as snow peas and french
beans favored the participation of small Guatemalan farmers at unprecedented rates (von
Braun etal., 1989; James 2000, Carletto et al., 1999). Other areas in the world,
particularly South America, Asia and countries in Africa such as Kenya, South Africa,
Senegal and Zambia are also quoted in the literature as recent success stories of
agricultural NTAX development (J affee 1999; Aksoy and Beghin 2005; Okello 2006;
Swinnen 2007).

The third boom has been recorded in the literature recently. This boom is about
the expansion of structured procurement networks such as those of supermarkets and
restaurant chains (the latter also referred to as the foodservice sector) within national and
regional markets. Because of their demanding quality, safety and delivery standards in
comparison to traditional markets, the market channels in this boom are named high
‘ standards supply chains (HSSCS) to differentiate them from the NTAX supply chains.
The development of these procurement systems has unveiled a number of opportunities
for small farmers (Granados 2004; Berdegue et 1a., 2005; Balsevich 2006; Hernandez et
al., 2007; Reardon and Flores 2006). Notwithstanding, despite being locally based and
offering higher income opportunities for small farmers, a systematic exclusion of small
farmers from HSSCs is feared (Berdegue et al., 2005). HSSCs impose strict compliance
standards of quality, food safety and delivery conditions as a need to compete with other
market channels, avoid potential problems with high end consumers and comply with
corporative product standards such as the case of McDonalds and Pollo Campero, two
multinational companies with over 400 outlets in Central America (personal interviews

2008). Central-America is not the only region where this phenomenon takes place with

similar market channel requirements. In other areas of the world, labor laws and, in some
cases, environmental protection requirements also need to be fulfilled in addition to

quality and safety standards (GlobalGAP 2008).

1.4 Research objectives

The major objectives of this dissertation are threefold. First, it investigated how
the roles of actors in the lettuce production sector in Guatemala evolved and the
interrelations between suppliers, service providers, support agencies and buyers in the
HSSC market channels. This was accomplished by conducting a value chain assessment
of the Guatemalan lettuce sector through use of key informant interviews and existing
knowledge on alternative value addition streams for farmers supplying retailers and
foodservice companies. Formal, peer-reviewed research on the subject does not exist for
lettuce or similarly perishable products in Guatemala; therefore, this phase of the
dissertation provided critical information about, among other things, the existing
networks across farmer organizations, distributors and retailers and respective supply
requirements. The value chain assessment also provides important scoping and context
information as the foundation knowledge for the other two dissertation research
objectives addressing the importance of asset endowments and limitations critical in
accessing HSSC market channels. Findings through key informant interviews and
secondary data for this research were collected in September 2008.

Second, an in-depth analysis of a successful small farmer organization was carried
out. The case study outlines the role of network assets in the organization’s management
structure that has facilitated access from local markets to higher value added, HSSC’s.

Contradicting views are often found in the literature on farmer organizations. For

instance, the literature from NTAX adoption suggests that farmers affiliated
withproduction and marketing organizations are more likely to leapfrog from traditional
crop production (e. g., basic grains) to more profitable market venues as they improve
their access to information, inputs, credit and technical assistance collectively (von Braun
et al., 1989; Carletto et al., 1999; Hamilton and Fischer 2005). Hewever, Hernandez et
al., (2007), described that such organizations do not necessarily provide specific
marketing services to connect small farmers with a broader network of wholesalers and
retailers in dynamic markets. To shed more light into this topic, this research targeted the
oldest, most successful lettuce producing farmer group, Labradores Mayas. Labradores
Mayas’ entrance to HSSC’s allows the comparison and contrasting of the organization’s
role in facilitating its members’ access to HSSC’s for nearly a decade. Information for
this research was obtained through several interviews with key Labradores Mayas
managers, members and their network of input, service providers and buyers during
October 2008.

The third major objective of this dissertation has been to investigate the combined
role of farmer household assets (physical, human, network and environmental) in
supporting farmers’ access to HSSC’s. To accomplish this objective the research departed
from the hypothesis that, all else equal certain threshold of household assets are necessary
for small farmers to enter and remain active in one or more of the HSSC market channels.
The research builds on the literature on determinants of market channel adoption,
particularly the works of von Braun et al., (1989), Barham et al., (1995); Carter et al.,
(1999) and Okello (2006) concerning NTAX, and Balsevich (2006) and Hernandez et al.,

(2006) on supermarket channel adoption.

An innovative contribution of this research is the addition of a set of network- and
environment-related assets not previously assessed empirically in the context of
determinants of market channel adoption. For instance, network assets such as number of
up and downstream linkages in the value chain were regressed, providing important
insights concerning their role in HSSCs access. Environment-related assets such as
propensity to frost and altitude were also included, the results of whichenrich the
literature and outline new gaps that need to be addressed concerning the role of
enviroment-fixed production conditions and compliance with key HSSCs requirements.
Primary data for this research was collected in Guatemala from November to December
of 2008 through a survey of 327 lettuce family farmers. The subjects were stratified by
whether they supply traditional markets orHSSC’s.

This dissertation is organized in four additional chapters. Chapter 2 details the
results of the lettuce value chain assessment. Chapter 3 describes the case study of
Labradores Mayas as the long-standing farmer organization in the lettuce value chain
with almost ten years of success supplying the supermarket channel. Chapter 4,
investigates the combined role of a set of farmers’ assets in gaining access to HSSC’s
using quantitative research techniques. Chapter 5 summarizes the three essays’ findings
into the broader development context with final considerations for development and
further research. All essays are followed by table summaries, support data and figures
with a final appendix at the end presenting field research protocols used for all three

essays.

References

Aksoy, M.A., Beghin, JC. 2005. Global Agricultura Trade and Developing Countries.
Business and Economics. World Bank, Washington, DC. .

Bardhan, Pranab. 2004. The Impact of Globalization on the Poor. Brookins Trade Forum.
271-284.

Berdegue, J.A., F. Balsevich, L. Flores, T. Reardon. 2005. “Central American
supermarkets’ private standards of quality and safety in procurement of fresh fruits and
vegetables,” Food Policy, Vol 30 Issue 3, June 254-269.

Balsevich, F., J. A. Berdegue, L. Flores, D. Mainville and T. Reardon (2003).
"Supermarkets and Produce Quality and Safety Standards in Latin America." American
Journal of Agricultural Economics 85(5): 1147-54.

Balsevich, F. 2006. Essays on producer’s participation in, access to, and response to the
changing nature of dynamic domestic markets in Nicaragua and Costa Rica. Doctoral
dissertation. Department of Agricultural Economics, Michigan State University.

Calvin, L., Foster, W., Solorzano, L., Mooney, J .D., Flores, L., & Barrios, V. (2002).
Response to a food safety problem in produce: A case study of a cyclosporiasis outbreak.
In B. Krissoff, M. Bohman, & J. Caswell (Eds), Global food trade and consumer demand
for quality (pp. 101-128). New York: Kluwer Academic Press.

Carletto, C. A. de Janvry and E. Sadoulet. 1999. “Sustainability in the Diffusion of
Innovations: Smallholder Nontraditional Agro-Exports in Guatemala. Economic
Development and Cultural Change. 47 (2): 345-369.

Granados, J .C., 2004. “Analis del Comercio Regional y Extragional de Frutas y
Hortalizas en Centro America”. PFID-F&V, Michigan State University. Consultancy
report non published report.

Hamilton, S. G, Fischer, E. F. 2005. Maya Farmers and Export Agriculture in Highland

Guatemala: Implications for Development and Labor. Latin American Perspectives,
2005; 32; 33 Latin

Hernandez, R., T. Reardon, and J .A. Berdegue. 2007. "Supermarkets, Wholesalers, and
Tomato Growers in Guatemala," Agricultural Economics, 36(3), May.

Jano, P., Mainville, D. 2006. Public & Private Roles in Promoting Small Farmers’ Access
to Non-traditional Markets: Case Studies from Central America. IAMA submission 2006
Conference. Buenos Aires, Argentina

James J .W., Glenn H. Sullivan, and Guillermo E. Sanchez 2000 “Future market
development issues impacting Central America’s nontraditional agricultural export

sector: Guatemala case study.” American Journal of Agricultural Economics. 82: 1177—
1183.

Kula, Olaf; J. Downing and M. Field. 2007. Globalization and the small firm: An
industry value chain approach to economic growth and poverty reduction. Micro Report
42. USAID.

Lamb, J. 1991. Final Report on Proexag Non-Traditional Agricultural Export Support
Project. Chemonics International on line records.

Maertens, M. Swinne, JFM. 2007. Trade, Standards and Poverty. Licos Centre for
Institutions and Economic Performance and Department of Economics, University of
Leuven, Belgium.

Maertens, M., Colen, L. and Swinnen, JF.M. 2008. Globalization and Poverty in Senegal:
A Worst Case Scenario? LICOS Centre for Institutions and Economic Performance. &
Department of Economics. K.U.Leuven.

Meller, P. 2008. From unilateral liberalization to regional free trade agreements: a Latin
America perspective. Econ Change Restruct (2009) 42285—103. Published online.

Okello, J .J ., and SM. Swinton 2007. “Compliance with International Food Safety
Standards in Kenya’s Green Bean Industry: Comparison of a Small and a Large Scale
Farm Producing for Export.” Review of Agricultural Economics 29(2): 269-285.

Outreville, J. F. 1996.Trade in Insurance in the Central America Common Market. World
Economy; Sep96, Vol. 19 Issue 5..

Pomareda, C. 2001. Small Farmers and their Participation in Central-American
Agricultural Exports. UNCTAD Regional Workshop Proceedings on the Regional
Integration and International Linking for the Agro Food Sector Development. San Isidro
de Coronado, Costa Rica.

Reardon, T. and L. Flores. 2006b. “ Viewpoint: ‘Customized Competitiveness’ Strategies
for Horticultural Exporters: Central America Focus with Lessons from and for other
Regions,” Food Policy, 31(6).

Swinnen, J.F.M. (ed.), 2007. Global supply chains. Standards and the poor. Oxford: CABI
Publishing. '

von Braun, J. Hotchkiss, D., and lmmink, M. 1989. “Nontraditional Export Crops in
Guatemala: Effects on Production, Income, and Nutrition,” Research Report no. 73

(International Food Policy Research Institute, Washington, DC, 1989).

Williams, Robert G. 1986. Export Agriculture and the Crisis in Central America. Chapel
Hill, NC., USA. University of North Carolina Press.

10

CHAPTER 2

A VALUE CHAIN ASSESSMENT OF GUATEMALA’S LETTUCE SUBSECTOR

2.1 Introduction

Linking small farmers to better market opportunities to increase their incomes
continues to be a major global concern for development scholars and practitioners. Yet,
constant changes in markets keep challenging how small farmers can be supported
through more effective, sustainable strategies (Kula et al., 2006).

The developing debate on how small farmers can access dynamic, higher paying
markets has been dominated by two lines of thought. On one hand, it is said that small
farmers can make a choice to opt for higher paying markets by systematically complying
with stringent quality, food safety, and logistics standards (Calvin 2002; Farina 2000;
Dries and Swinnen, 2004; Hu et al., 2004; Berdegue et al., 2005; Balsevich et al., 2006,
Hernandez et al., 2007). Other researchers argue that even when farmers can fulfill most
or all the market requirements, they still incur higher transaction costs than larger
suppliers, leading to a potentially systematic elimination of small farmer participation
(James 2000; Pomareda 2001; J ano et al., 2004; Gulati 2006; Okello 2006). The debate
goes on, calling for further evidence focused not only on recent failure or success stories,
but also on small farmer market penetration and management models that have helped
them remain in business over the years (Cropp 1989; Adesina and Djato 1996; Swinnen
2005).

To add evidence to this wide-ranging topic, this paper developed a value chain
assessment of the lettuce subsector in Guatemala to provide a retrospective, yet

crosscutting, view on how small farmers can successfully and sustainably access more

11

profitable, high standards supply chains (HSSCs). “Value chain” means the full range of
activities and services a company performs (either vertically integrated or coordinated
with suppliers and service providers) to bring a product to its final step—the consumer
(Kula et al., 2007; Kaplinsky and Morris 2000, Feller 2006). The detailed information
gathered through the value chain assessment provides useful insights and ideas from
leading farmer groups on how markets can be put to work in favor of poverty reduction.
Value chain assessments have become popular in international development
practice as a means to provide small and medium-scale firms strategies to strengthen their
linkages with up and downstream actors thereby keeping them apace with changing
market conditions (Kula et al., 2000; OECD 2008). For instance, empirical evidence
from cooperatives and contract farming research shows the important role of trust and
coordination between small farmers and service supplier and buyer links to help them
reduce transaction costs (Cheung 1983; Williamson 1985; Hart 1995; Key and Runsten
1999). In value chains, analogous dimensions to cooperatives and contract farming in
terms of trust and coordination between up and downstream links represent the vital
bonding for it to work (Feller 2000, Downing and Kula 2007; Webber and Labaste 2009).
Based on agency theory (AT) and transaction costs economics (TCE) this paper addresses

the following research questions:
0 How were lettuce HSSCs networks established in Guatemala?

0 What are the roles of trust and coordination across the value chain for small

farmers to access HSSCS‘?

12

The study focuses on the lettuce sector in six communities in the Guatemalan
Highlands (Figure 2.5). It was confirmed during interviews with key-informants that
these communities are the primary fresh lettuce sourcing area for HSSCs and the
domestic and regional traditional market. There are two critical characteristics of the
farmers under study concerning challenges to access HSSCs: First, farmers in the study
are small-scale producers with average land tenure ranging from less than 0.5 hectares up
to 2.5 hectares (Hamilton and Fischer 2003 and 2005). Consequently, nearly 50% of
lettuce suppliers in this area plant, tend, and harvest lettuce on plots averaging 1/10 of a
hectare. Second, despite the complex production, post-harvest handling, and delivery
logistics of this product in comparison to less perishable vegetables such as carrots or
cabbage, small farmers’ share of sales to HSSCs has grown steadily in the last 15 years.
This length of time accessing HSSCs provides an adequate window of analysis to identify
how the farming networks were established and how trust and coordination among
groups have played a role.

This paper continues with a historical account of Guatemalan agrarian systems
with small farmers and a general description of the lettuce subsector. Section 2.2
summarizes the existing theoretical background on value chain assessments and its
connection with the theory on agency and transaction costs. Section 2.3 discusses the
field methods used for the study, while section 2.4 summarizes the information obtained
from interviews and secondary data in response to the research questions. Section 2.5

presents conclusions of the value chain assessment.

13

2.1.1 Guatemalan small-holder agrarian systems

The agricultural development context in Central America is tightly linked to land
ownership and the persistence of economic inequalities. For instance, land ownership has
been the foremost asset associated to wealth accumulation and distribution in
Guatemala’s colonial and post-colonial history (Williams 1986; Davis 1988). In the last
decades, those with the largest land allocations have accumulated even more land
throughout several agricultural development booms (Williams 1986). As a result,
Guatemala has one of the most unequal land-distribution Gini coefficients (0.85), which
is among the highest in the world (Green 1989; Thery etal., 1988 quoted in Barham et
al., 1994). The production booms of bananas, cotton, sugar cane, and cattle have
historically enabled large land holders to expand in size, while small farmers have
continually worked the least fertile and hilly lands of the piedmonts and highlands.
Consequently, one export boom after another contributed to the consolidation of a dual
agrarian system (large-scale versus subsistence farming) some benefitting large farmers
but some benefitting also small farmers (Barham et al., 1994). The history of one boom
after another is outlined in the literature has positive for the agricultural sector as it has
arguably generated the local expertise to continuously participate in new, better market
alternatives.

One of these booms, supported first through donor programs was non-traditional
export crops (NTAX) targeting off-season and niche markets in Central America, the
United States and Western Europe. Even though small farmers were not specifically
targeted as beneficiaries of the NTAX production and market research, small agrarian

systems had particular advantages to participate (Lamb 1991). Not all who tried NTAX

14

became success stories, and not all success stories in the NTAX boom favored small
farmers (AGEXPORT 2008). Nonetheless, those crops that were well-adopted by small
farmers in the highlands had the following characteristics: (1) Labor-intensive fruits and
vegetables such as raspberries, blackberries and snow peas were well adopted by small
farmers, since labor is an abundant element in the highly populated rural areas of
Guatemala. (2) Short-cycle products grown more than once a year in the same land. (3)
Higher market value in comparison to corn and other staples, which allowed farmers to
increase household incomes at a faster rate. (4) The adaptability of the crops to different
microclimates found in a relatively small geographic area in the highlands (James 2000;
Hamilton and Fischer 2002; AGEXPORT 2008).

Since the 19805 Guatemala’s NTAX experience has been often quoted in the
literature as a success story. Most participating farmers view NTAX as a means for self-
advancement and as an opportunity to use small plots of land and family labor (James
2000; Hamilton and Fischer 2002). Addressing indirect effects of NTAX, Helling et al.,
(2005) documented how farmers felt that in many ways the production of NTAX has
preserved affective ties to the community and reinforced key elements of the indigenous
cultural heritage. Other literature points out that the opportunities for smallholders
brought in by the production of NTAX have also provided continued training for small
farmers to access alternative market channels, both domestically and regionally (Reardon

and Flores 2007).

2.1.2 Guatemalan lettuce subsector

The production of lettuce in Guatemala is concentrated in the highlands where the

agro climatic characteristics allow year-round production of temperate-weather crops.

15

The highest concentration of producers for traditional and HSSCs is clustered in the
departments of Guatemala, Sacatepequez and Chimaltenango, with other clusters of
growers in the departments of Solola and Quetzaltenango (INE 2003). These areas
produce most of the lettuce consumed in Guatemala and El Salvador, with a small
fraction shipped to Honduras a few months a year. El Salvador and Honduras have only a
few areas with ideal agroclimatic conditions to grow lettuce year-round (Granados 2004).
A recent national survey of the agricultural sector reports about 1400 small and
medium farmers growing lettuce in Guatemala (National Statistics Institute — INE -
2003). The survey shows that most Guatemalan lettuce producers in general own less
than 3.5 hectares. Triangulated information between area planted, trade data for the
Central American market and secondary data on the fruit and vegetable sector point out a
total country lettuce output of circa 13783MT (INE 2003; Granados 2004; CENMA
2007, MAGA 2007). The lettuce value chain map for 2003—2004 is presented in Figure

2.3 and estimated production from several sources in shown in Table 2.7.

2.2 Theoretical basis for value chain assessments

The value chain notion is nested on business and management theory. As such, it
becomes a practical tool to understand the relationships of the small and medium-sized
enterprises with its backward and forward linkages and networks with the purpose of
generating recommended actions (OECD 2008). Porter (1985) made the notion of value
chain development known to practitioners and scholars through his research on firm
performance analysis to increase competitiveness. Through this analysis, a firm can
identify where to improve primary activities such as Operations and logistics that have a

direct impact on the value created to the customer and the margin to the firm. The firm

l6

can also analyze improvements to support activities such as human resource management
or technology development that impact value to consumers and margins to the firm by
increasing business performance (Figure 2.4).

To Porter’s credit, the concept of value chain has evolved into arguably the most
widely used paradigm to understand and formulate alternatives for firm and industry
development (Ankli 1992). In the last ten years, a growing number of development
practitioners and scholars have also adopted the value chain paradigm as a framework to
analyze the links where small farmers can become more competitive. The value chain
notion in an international development context overlaps with issues largely studied in
economics particularly concerning agency and transaction costs. Consequently, business
theory meets economics and development theory when pro-poor industry value chain
assessments are expanded beyond the firm to other firms upstream and downstream in the
value chain.

In international development, interlinked relationships such as contract farming
and cooperatives have been extensively studied to understand how inputs and services are
brought together for famers to grow, transform, and market products to consumers
profitably and sustainably (Cheung 1983; Williamson 1985; Hart 1995; Key and Runsten
1999; Kristen 2002; Bagetoft and Olesen 2004; Webber and Labaste 2009). The
additional contribution of the value chain framework, as Webber and Labaste (2009)
explain, is that the value chain “sheds light on the size of the firms participating in each
link, how they are or could participate in the chain and provides an analysis of the
opportunities to facilitate or improve those linkages.” Therefore, the value chain

approach is useful in further dissecting and understanding the relationships from link to

17

link to the consumer while outlining areas of improvement or defining business
development services targeted to the small and medium enterprise (OECD 2008; Kula et
al., 2006; Webber and Labaste 2009).

Analytical frameworks parallel to Value chain assessment at the firm level
include the supply chain analysis, while other—not mutually exclusive—approaches are
more concerned with overall industry performance such as the filiere (French word for
connections or network) and cluster assessments. In the first case, Feller et al., (2006)
present one of the clearest descriptions of supply chain as “the activities concerning
logistics and procedures involved in the production and delivery of goods and services
from procurement of inputs to the fulfillment of customer’s satisfaction.” In other words,
the supply chain is the set of activities involved from the supplier’s supplier to the
customer’s customer, with the highest possible efficiency (Feller et al., 2006). In the
second case, industry-oriented analytical frameworks have been used to guide
government and donor programs to improve the business environment of target sectors
and subsectors. The filiere industry analysis and the cluster assessment both continue to
be widely used by government and donor programs supporting national-level
competitiveness of specific sectors and subsectors (Webber and Labaste 2009). Their
origins, however, date back to the early-and mid-19605 when understanding and
facilitating vertical and horizontal coordination among companies favored economies of
scale, encouraged contract farming, and improved overall industry performance (Cook
2000; Sykuta and Cook 2001).

Empirical research from China and India points out that AT and TCE are

important in value chain assessments as both depart from the assumption that people are

18

opportunistic in seeking self-interest rather than cooperating (Becerra and Gupta, 1999,
Chow 2008). Following this theoretical basis two propositions are tested in this paper to

illustrate the relevance of AT and TCE theories in value chain development.

Proposition One (agency theory):
Trust within the value chain is built on positive past experience between links thus

constituting the major network building blockforfarmers accessing HSSCs.

Williamston (1991) stresses that high levels of trust lead to cooperation that in
turn can lead to lower transaction costs and fast response capabilities. These are ideal
conditions in value chain development, thus allowing two or more links to build and pass
forward a higher value proposition to the consumer (Hill 1995). Consequently, the value
chain becomes a collection of networks working together from link to link under a higher
level of trust (Feller 2000; Downing and Kula 2006). The upshot of higher levels of trust
in the value chain is the reduced inclination of participants to watch for opportunistic
behavior (Zaheer, McEvily, and Perrone 1998). Additionally, trust is a good indicator of
how individuals are encouraged to form networks of collective technology learning, risk
handling and information sharing with their immediate link downstream (Smallbone and
Welter 2001). The issues of opportunism and trust in the development of inter-firm
business relations have been extensively studied inside and outside the value chain
framework. Nooteboom (1996) established that even the mere perception of
trustworthiness of partners can reduce transaction costs. As such, access to one network

(for example, in a horizontal coordination scheme in the value chain) sends a strong

19

message to partners in another network (for example, a downstream horizontal level, or
higher vertical coordination link). This trustworthiness builds reputation capital that can
facilitate network expansion in both directions of coordination, thereby reducing

transaction costs (Nooteboom 1996, Reardon and Flores 2007).

Proposition two (transaction cost economics):
Asset-poor farmers are more likely to behave opportunistically in complying with the

minimally required value chain functions than better-ofl farmers.

TCE explains key dimensions to understand farmers’ limitations that foster
opportunistic behavior in commercial relationships. When farmer characteristics (or
assets) constrain them from overcoming imperfect input and credit markets, adopting the
necessary technology, accessing information, and managing risks, the capacity to
improve value and profits becomes more difficult (Eisendhardt, 1989, Bromiley and
Cummings, 1995, Okello and Swinton 2007). Several authors have categorized these
asset limitations based on empirical research since the mid 19805: (1) physical assets
concerning farm size, machinery, and other production-specific physical assets; (2)
human assets concerning age, education and experience of the entrepreneur; (3) temporal
assets concerning the geographic location and conditions of the farm to deliver goods and
services within a timeframe that minimizes costs and increases value (Martinetz 2002,
Okello and Swinton 2007). These assets have also been assessed in a similar context to
evaluate TCE (Runsten 1992; Warning and Key 2000; Ponte 2000, Okello 2006;

Balsevich 2006; Hernandez 2007).

20

How AT and TCE affect different market channels can be further hypothesized
based on the expected coordination between links to comply with tangible and intangible
product traits in HSSCs. For example, different arrangements are found between links
when comparing lettuce HSSCs under the value chain approach with the traditional
market channel. In value chains, participants in one link are bound through network
arrangements including contract farming or outgrower schemes—implicit or explicit—
with links upstream, downstream, or horizontally. This also assumes that all links are
synchronized with the end market where the product competes on cost, quality, volume
and continued reputation (Reardon and Flores 2007). Value addition starts from accessing
information on procurement of adequate and lower-cost inputs to improving production
technology, desirable product characteristics, logistics, and other product features (Feller
2000, Downing and Kula 2006). In comparison, suppliers in the traditional market
channel are less influenced by competing on tangible and intangible product traits
(Berdegue et al., 2005; Reardon and Flores 2007). As such, the expected price for a
product is less connected to the notion of value proposition to the consumer. This is
because the spot market does not promote differentiation or brands and does not feedback
information that producers can use to improve value. Additionally, traditional markets
rely on minimal market failure, immediate access to price information, and short-term
performance (Webber and Labaste 2009). Using AT and TCE theory to test these

propositions will further develop this framework.

2.3 Value Chain Assessment Methods

The methodology for the value chain assessment was adapted from Value Chain

Handbook (Feller 2000) and with insights from Kula et al., (2006). Kula et al., (2006)

21

describe how value chain analysis furthers subsector analysis on four key points: (1)
inter-firm cooperation from suppliers to wholesalers and down to retailers to enhance
competitiveness; (2) power and trust relationships across actors that can increase
collective efficiencies; (3) distribution of risks and benefits as the dynamics that create
upgrading incentives or disincentives; (4) understanding that learning and innovation by
small farmers is important to explain how they gain access to new skills, and adapt to the
changing market conditions on a continuous basis.

The link between the research questions and the methodological approach lies in
an extensive open-ended interview around these four issues with the goal of obtaining
information on the existing support networks throughout the production and distribution
links of the value chain. Open-ended questions are preferred in value chain assessments
so that interviewees talk freely about context—rich personal experiences in production and
marketing (Feller 2000).

For this study three stages of data collection were carried out. First, an extensive
literature review on the lettuce subsector in Guatemala provided important qualitative and
quantitative data on production volumes, market share of different types of
intermediaries, and direct supplies to HSSCs. Second, key informant interviews were
conducted with relevant representatives of the lettuce value chain. The selection of the
informants was done making sure all sectors were represented. The third stage was done
through secondary interviews with key people where clarifications and validation of data
was necessary. Data triangulation was done exhaustively across interviewees following
qualitative method recommendations by several authors (Maxwell 1996; Fielding and

Fielding 1993, Rubin and Rubin 2005).

22

Among the people interviewed were rural farmers with and without membership
in a small farmers’ organization with lettuce being the major crop produced. The data
collection took place in the summer of 2008. A total of nine farmers were interviewed at
different depths, based on their experience. From these nine, three were farmer
organization leaders, two were founders and one was a hired support professional. A total
of nine intermediaries that gather lettuce in key gathering areas were particularly targeted
to understand the supply networks to the Salvadoran market, either on a year-round basis
or during specific seasons. From these intermediaries, five also supply the Guatemalan
traditional and HSSCs markets and one supplies the restaurant chains. Purchasing
managers of the first- and second-tier supermarkets in Guatemala City were also
interviewed. Several interviews with support institutions were scheduled, but only four
NGO and one Ministry of Agriculture representatives were completed. At least three
input providers (supplying seeds, pesticides and fertilizers) were visited in the zone

without complete interviews, but important support information was gained nonetheless.

2.4 Findings
2.4.1 The lettuce value chain networks

Since the advent of modern transportation and fast urbanization in Central
America, Guatemala’s central and western highlands have been the major supplier of
temperate weather fruits and vegetables for the region (Granados 2004; SIECA 2008).
Four distinguishable market channels have originated since the late 19705, which are
regrouped as traditional and HSSCs market channels later in this analysis. Understanding
the history associated with each of these market channels sets the tone to dissect the

subsector structure, while the detailed assessment of participants, market shares, and

23

buyer-supplier relationships provides evidence in support of the theoretical propositions
under evaluation. Additionally, the breakdown of actors also illustrates the importance of
network building as a key element of success for small farmers involved in one or more

market channels.

The Guatemalan Traditional Market

Guatemala City has become the major trading spot for highland and low valley
fruits and vegetables with daily transactions in 2008 nearing US$05 million a day
(MAGA 2008). The ramifications of such dynamic trade have been extended to urban
and rural markets as far north as Chiapas in Mexico and as far south as Costa Rica
(SIECA 2003). Trade data from SIECA and the Ministry of Agriculture report different
values of fresh fruits and vegetables traded in the Guatemalan traditional market
averaging over $170 million in 2006 in just two major wholesale points, La Terminal and
the Wholesale Center (CENMA). This market channel also provides the meeting point
for wholesalers from El Salvador and nearly 150 wholesalers supplying the city markets
of over 300 towns and villages in Guatemala, El Salvador, and, during some seasons, also
in Honduras (CENMA 2008). Based on interviews with wholesalers in the traditional
market and consulted secondary trade data, the market share of this market channel from
total lettuce production in 2007 has been calculated at 35% (circa 4100MT), and is grown
mainly by small farmers (Figure 2.4).

Despite the existence of business relationships between the most experienced
wholesalers and supplier for over three decades, business transactions still take place as a

spot market. Two criteria dictate the value of the product: immediate verification of

24

tangible quality and the day’s set price. For suppliers, little has changed over the years in
terms of business relationships, while wholesalers continue to dictate prices throughout

the season.

The Salvadoran Traditional Market

Following a devastating civil war in the late 19705 and early 19805, El Salvador
has experienced over two decades of peace and economic development. By 2004 San
Salvador has grown into a modern metropolis of nearly three million people, creating
immediate demand for warm- and temperate—weather fruits and vegetables (Granados
2004). Improved economic times also resulted in the fast proliferation of rural markets in
western and eastern El Salvador, where purchasing power has been boosted by a growing
level of remittances from Salvadoran immigrants in the US to their Salvadoran families
back home. In 2006 alone, El Salvador received over $2.6 billion in remittances,
accouting for 17% of the GDP. By 2008, the Salvadoran Central Bank estimated that
remittances totaled $3.8 billion, with at least 22% of Salvadoran families receiving
remittances (US State Department 2009).

Supply opportunities created by a rising demand for fruits and vegetables could
not be tapped by El Salvador’s production capacity (Granados 2004). Most of the
agricultural production land in El Salvador is in the low valleys along the Pacific Littoral,
and only a few areas have the necessary agro-climatic conditions to grow temperate
weather crops. Guatemala’s diverse climates and longer history of experience with
diversified agricultural production became the natural solution to this demand. Warm-

weather and temperate-weather vegetable crops are shipped to El Salvador from areas

25

within 350km of Salvadoran border including from Guatemala’s western highlands
(MAGA 2008). Although several Guatemalan wholesalers have supplied this market
since the years of civil war, interviews point out that Salvadoran traders coming with
trucks to Guatemalan territory were the pioneers of this important market channel.
Working on a cash basis, most Salvadoran wholesalers have a purchasing model based on
three alternatives: (1) collection and payment at the farm gate or by the road; (2)
collection and payment at the terminal markets in Guatemala City or regional spot
markets (such as specific parking lots in city centers or gas stations); and (3) purchase of
full plantation based on yield appraissals days before harvesting (a common practice in
the eastern part of Guatemala for onions, tomatoes, peppers, cucumbers and
waterrnelons). According to interviewees these alternatives have not generated long-term
relationships between farmers and wholesalers. Repeated opportunities to engage in
transactions are governed by spot-quality assessments. Despite the continued affluence of
Salvadoran buyers for the past thirty years, positive past experience has not motivated a
different arrangement such as contract farming, exclusive supply contracts, cost-sharing,
or production land acquisitions.

On the other hand, some improvements are quoted to the way transactions are
made. In the last ten years, telecommunications and the availability of banks in
practically every town have favored the adoption of modern banking practices. For
example, the use of the banking system to avoid handling large amounts of cash in the
field is a common practice for over half of the seventy permanent wholesalers buying
lettuce in the Central Highlands. Even if these are important changes, tracking how this

market channel introduces change is difficult, as most participants operate in an informal

26

economy with small-scale suppliers. Additionally, most interviewees calculated that at
least seventy permanent wholesalers and an equal number of seasonal wholesalers trade
with five hundred small farmers along the roads and inland production spots of the
Western and Central Highlands. Cross-checking secondary trade data and census data it
is calculated that 15% of 2008’s total lettuce production (circa 1750MT) grown by small
farmers was traded through Salvadoran wholesalers in this channel.

Guatemalan traders also supply the Salvadoran market channel. These traders
have experienced the most changes within the traditional market in almost three decades
of experience. Most of them have gone from taking a small truckload per week to spot
markets in the early 19805 to taking weekly shipments of 40-foot, refrigerated containers.
Consistency in variety, quality, and quantity has helped these firms to reach an
understanding of open market dynamics to the point where spot sales have become
prearranged deals with specialized wholesalers catering to stores, hotels, and restaurants.
In the produce business “deals” is a term for expected harvests of specific products. The
term is now used by large wholesalers in the Central America produce jargon. A clear
example of a deal is “the southern Texas melon deal” or “the Chilean deal” for several
products (Lamb 1991). Additionally, most of the Guatemalan wholesalers are no longer
operating in the informal economy, but have matured into fully coordinated enterprises
with outgrower schemes, large collecting, sorting, and packing facilities. At least two
have their own distribution facilities in El Salvador (Disvegua 2008). While 23% of the
total share of lettuce grown in Guatemala is sold in El Salvador, 8% of this volume (circa

1040MT) is traded by a handful of Guatemalan wholesalers (Table 2.1 and Figure 1).

27

As Guatemalan wholesalers supplying this market channel have grown in scale
and sophistication, it is not yet clear how the issues of trust affect possible network-
building schemes. It is for future research to watch the evolution of these actors and the

value chain development models they adopt to see whether they favor small farmers.

The supermarkets channel

The history and evolution of supermarket procurement systems in Central
America has received close attention in the literature in recent years. Berdegue et al.,
(2005) describe how supermarket procurement systems have changed from a local store-
by-store system to the building of distribution centers (DCs) following the development
patterns of similar companies in developed countries. This market channel has undergone
a period of fast expansion with deep structural changes, particularly in the areas of
consolidation and multinationalization (Berdegue et al., 2005; Reardon and Flores 2007).

Central America is home to about thirty-five million consumers, of whom roughly
a third live in the capitals and second-largest cities, making it an important trade corridor
with a purchasing power similar other important Latin-American markets like Colombia
or Argentina (MAGA 2008). For multinational retail companies, unifying trade standards
and increasing efficiency across borders were two of the major obstacles that needed to
be addressed to spur business opportunities in the region. These obstacles have been
gradually surmounted through multilateral efforts to standardize customs procedures for

faster cargo transit dating from the 19605 (Lamb 1991). An efficient customs union

28

system has been put in place, thanks in part to the dispositions recently signed under the
Central America and US Free Trade Agreement (CAFTA-DR').

Nearly six years before CAFTA-DR was signed, the Dutch company Royal Ahold
ventured into retail investments by buying 33% of the Guatemalan company La Fragua
and the Costa Rican company CSU in 1999. Family-owned and with seventy-five years
of experience catering to the Central American consumer, these two companies had by
1999 established nearly 375 outlets in Guatemala, El Salvador, Honduras, Nicaragua and
Costa Rica. Their forte became the use of different store formats that customized
products and services according to income class and level of urban development in
capital cities and the rural areas (Berdegue et al., 2005). Riding on these companies’
experience and formidable past performance, Royal Ahold began a new company
immediately baptized as the Central American Retail Holding Company (CARHCO).

The following years of CARHCO saw an aggressive expansion of stores, as well
as the addition of two different formats targeting popular and rural areas across Central
America (see Table 2.2 and 2.3 and Figure 2.2). As this development took place, new
procurement policies and practices were introduced aiming at gaining market share from
traditional village and neighborhood retailing models and La Torre supermarkets (the
second-tier supermarket company in Guatemala). Troubled finances at Royal Ahold’s
home base in the Netherlands at the end of 2004 opened up the opportunity for Wal-Mart
to buy Royal Ahold’s share in CARHCO. Months after acquiring Ahold’s shares, Wal-
Mart purchased 51% of the holding in 2005. CARHCO became Wal-Mart Central

America and further growth followed. By mid 2008 the number of formats was increased

 

' CAFTA-DR was signed in 2007 and also includes the Dominican Republic
29

to twelve and the number of stores to 520, of which 249 are located in Guatemala and El
Salvador (Table 2.1, Table 2.3 and Figure 2.2).

In the midst of the changing winds of modernization of the retail sector, La Torre
sought to compete by emulating the structural changes adopted by its long-time market
nemesis in Guatemala, La Fragua. La Torre merged with Econosuper under a new
company name, Unisuper, in 2001 (see tables 2.2 and 2.3). La Torre owns forty-five
stores with two major retail formats designed and adapted over time based on consumer
traffic and income level of urban neighborhoods. Expansion into rural areas has been
slow, but the growing trend points in that direction according to purchasing managers.

The procurement model of supermarkets has important implications for small

holders, which are analyzed in great detail in the next section.

The foodservice market channel:

By the early 19905 increasing numbers of restaurant chain outlets represented a
niche market for a variety of fruits and vegetable growers and wholesalers. Parallel to this
growth, wholesalers already supplying the growing Guatemala and Salvadoran market
first tapped the opportunities in the foodservice market. The procurement model
consisted of sorting the best quality available in the traditional market in Guatemala City
as a one stop shopping for all tropical and temperate weather products (personal
interview with leading wholesalers 2008). Once volumes of locally sourced vegetables
such as tomatoes and lettuce rose over 2MT/week, it became necessary to homogenize

quality and regulate delivery to the highest efficiency standards. By 2006, McDonald’s

30

consumed an average of 14MT of lettuce and a similar amount of salad tomatoes per
week (El Periodico 2006).

Although it was not outlined in the literature, the foodservice sector introduced
two major innovations to the fresh—produce procurement system in Central America
before supermarkets did. First, by 1990 both McDonald’s and Pollo Campero had
abandoned the traditional market as their source of produce, replacing traditional
suppliers with specialized wholesalers and direct grower/suppliers. To become a
preferred supplier, a year-long process was—and still is—necessary where compliance
with the Hazard Analysis and Critical Control Points (HACCP), a food safety assurance
system required. Second, a strong social responsibility program by both McDonald’s and
Pollo Campero has had rippled effects on the procurement system besides food safety
assurance to the consumer. For instance, it was pointed out during interviews with
farmers that general labor laws and compliance with no child labor-hiring are verified
and worker safety conditions and compliance with environmental protection standards are
continually reinforced through unannounced audit visits twice a year. Compared to the
foodservice market channels under study, it has taken supermarkets eight more years to
start adopting a procurement model that ensures safety and quality and fifteen years to
totally eliminate the reliance on traditional markets.

Tables 2.2 and 2.3 present a time sequence on the development of modern
procurement systems for the firSt- and second-tier supermarket and main restaurant
chains in Central America while Figure 2.3 shows the required tangible and intangible

product characteristics.

31

Both supermarket and foodservice channels compose the HSSCs market category.
In HSSCs, particular attention is placed on outlets in Guatemala and El Salvador as the
major destination of the lettuce grown in Guatemala. Exhaustive secondary data analysis
and interviews with suppliers to these chains point out that the foodservice market
channel consumed 6% of the total annual output of lettuce in 2007, equivalent to about
700MT. Although the major suppliers to the foodservice channel are organized small
farmers, interviews point out that there the vast number of restaurants not organized
under international chains source from wholesalers and from the traditional markets.
Tables 2.2, 2.4 and Figure 2.1 provide a detailed illustration of the structure and market
share of the participants based on information collected from key interviewees, trade
data, and newspaper articles on the subject. This is the first attempted effort to map the

lettuce value chain with this level of detail.

2.4.2 Differences in farmers trust and coordination across market channels

Gaps between traditional markets and HSSCs

A gap analysis in a value chain outlines the major strengths and weakness of one
market channel over another in terms of trust and coordination needed within farmer
organizations to comply with HSSCs added value generation. In this regard, traditional
markets suffer several disadvantages compared to HSSCs. For instance, traditional
markets are located in open-air spaces, vendors are assigned limited space to carry out
transactions; product is comingled, sorted, repacked or loaded from truck to truck under
no enforcement of sanitary conditions; there is no controlled access of people traffic,
rodents, birds, and other potential sources of contamination. To the small farmer it means

that no matter what production protocols he follows, the consumer will not know the

32

product came from his farm. Thus, unless the product looks bad superficially, there is no
punishment nor reward for product differentiation. HSSCs, on the other hand, have
abandoned this procurement system for a more coordinated supply chain model where
supplying small farmer orgnanizations must provide the minimum quality, cost, volume,
and efficient logistics required. The rewards for the extra efforts are in more than one
form. To illustrate, it was established that HSSCs paid on average $0.26 more per
kilogram in 2007 which has become an important incentive for small farmers to follow
rules. Even if complying with HSSCs standards creates higher costs, farmers still made
7% more profit on average per hectare planted than traditional market suppliers. Tables
2.5 and 2.6 detailed this information in greater depth.

Another key gap outlined in Figure 2.3 concerns product food safety and
traceability compliance programs. Without daily inspection of potable water and
approved pesticide use, there is ample window for small farmers to behave
opportunistically. However, continued access from 1995 to McDonalds and 2001 to
supermarkets shows small farmers in general are concious of the consequences of failing
to comply with these requirements. Why cannot traditional markets offer the same food
safety and traceability as do modern procurement systems? To explain this gap, four
reasons are outlined. First, traditional markets operate under fragmented supply chains
where knowledge of the origin of the product and the final consumer is not known and is
not relevant. The example of the practices used by Salvadoran wholesalers illustrates the
point. Second, price is set by spot recognition of tangible quality traits and imperfect
information on the volume dealt per day, per week and general notions on how volumes

will behave like in the season leaving little incentive to ensure intangible product

33

characteristics that are not verifiable at the spot market. Third, the lack of cold chain
facilities accelerate the speed of transactions in favor of the buyer as quality diminishes
by the hour in open air and the product has to travel immediately to the next selling point,
some times as far as 6-8 hours to El Salvador. This, in turn, illustrates that traditional
market channels are far from assuring similar shelf-life to the product as do HSSCs
suppliers. And fourth, traditional suppliers and buyers at La Terminal and CENMA2
cannot be liable for potential consumer contamination if there is no form to trace back
tainted products back to them. As transactions in traditional market channels take place
under an informal, cash-based economy, paper trail does not exist to document volumes
and terms of sale. Because of this situation, traditional suppliers lack the incentives (e.g.,
higher profits through higher value propositions) and capacities (e.g., physical
coordination with upstream links) to comply with required HSSCs standards. HSSCs
suppliers, on the other hand, have a reputation to build and take care of, favoring better
coordination with links downstream and strengthening trust bonds after continued
positive experience season after season. 5

Key interviewees pointed out that the need to presort (and, in most cases, pre-
wash) products sourced from the traditional market destined to supermarkets and
restaurants in the late 19805 marked the beginning of the formation of modern value
chains. While some traditional market agents pioneered the presorting, washing, and
delivering services in an efficient manner, the capacity to comply with climbing
requirements from downstream links beyond superficial characteristics was beyond their

reach. As a result, complying with superficial product characteristics became necessary,

 

2 La Terminal was the first traditional market location and is located near the center of Guatemala City.
CENMA was established in 1999 in the outskirts of the city as a measure to decongest truck traffic in La
Terminal.

34

but far from sufficient for all HSSCs market channels. Figure 1 describes the tangible and

intangible product traits required in HSSCs compared to traditional markets.

Gaps within HSSCs market channels

One of the major gaps within channels has been the late adoption of stringent food
safety assurance systems by supermarkets compared to foodservice. The question of why
supermarkets lagged in adopting more modern procurement systems compared to
foodservice can be answered by two major factors. First, during recent interviews with
key suppliers to both channels, it surfaced that foodservice companies have more at stake
when food safety scares have happened. For instance, the fatal case of Jack in the Box E-
coli contamination in the Seattle area in 1993 became every restaurant chain’s nightmare
ever since. In this unfortunate event four children died and hundreds of consumers
became ill (www.aboutecoli.com 2007). This was a wake up call to international
restaurant chains, leading to earlier adoption of stricter corporate policies concerning
food safety and quality since the early 19905. As became evident during interviews, the
relationship between small farmers supplying the foodservice sector became tighter as it
encompassed more explicit terms of the contract concerning audits from production to
delivery practices. The group Horticultores Unidos started selling to McDonalds in 1995,
and eversince their producers have remained compliant. This shows how opportunities
for higher profits and long-term relationships motivate trust and coordination over a long
period of time. Higher profits and rising weekly orders over time have allowed
Horticultures Unidos better planning in the field and longer-term relationships with its

member base.

35

The modern supermarkets have been for a long time companies that practically
rent space for other brands to be exposed and sold to consumers. This is changing with a
new wave of supermarket brand awareness in the last ten years and with the adoption of
private labeling programs that worked well in the US since the early 19905 (personal
inteviews with retailers). In respect to tougher purchasing standards of fruits and
vegetables one interviewee expressed the following: “Supermarkets were late in adopting
similar food safety standards as McDonald’s because they did not feel their brand and
company reputation was at the same level of risk compared to restaurants. You may buy
at Wal-Mart or La Torre, and you can find four or five brands of lettuce and none of them
say ‘Store’s choice’ as they do with some of their private-labeled goods. I don’t think we
will see supermarkets taking that risk with fresh produce.” This was a major gap until
2001 that supermarkets embarked in field inspections and verification of compliance with
minimum food safety and quality standards from the field. The lagged response was also
due to the nearly 200 different choices of produce among different weight presentations
and varieties supermarkets deal with. The high number of products, some sold in small
quantities like fresh herbs, some in large amounts like mangoes and melons, hindered the
ability to switch to direct suppliers able to comply with the minimum food safety and
traceability requirements.

From all trends in the market, assuring the safety of products has had the most
profound repercussions on how small farmer groups choose members, access HSSCs and
strategize to remain in those channels. Member reliability in following procedures is
directly linked to trust and coordination to avoid the farmer group management daily

verification visits to each production location. Table 2.6 summarizes the perception of

36

interviewees on the effects supplying and buying links have on different farmer assets,
particulary network assets, as key factors in entering HSSCs successfully.

A point common across several interviews that has direct connection to issues of
trust and coordination is the sense of pride small farmers feel feel when they gain
entrance to HSSCs market channels over those still supplying traditional market
channels. Small farmers summarized they felt better off with HSSCs despite the extra
effort it takes to remain in the market. Figure 2.3 shows how different business practices
are undertaken to allow compliance with tangible and intangible requirements. This
figure also shows the clear distinction for farmers between opting for HSSCs or staying
in traditional markets. This pride reflects satisfaction on how business is carried out with
HSSCs and the benefits that implies. In summary, farmers said that weekly fixed prices in
HSSCs were better than daily-set prices in the traditional market (Table 2.6). The
possibility to grow market share was another highly emphasized benefit in HSSCs, a fact
they had observed on a weekly and yearly basis with HSSCs and expanding trade
opportunities in the region. To cite one example, one of the leading suppliers to Wal-Mart
stores in Guatemala and El Salvador entered into a supply deal with Wal-Mart Costa Rica
supermarkets back in 2006 (personal interview with Labradores Mayas managers).

In support of propositions one and two, these gaps outline how AT and TC E
issues in rapidly changing market channels such as HSSCs are interwined with the small
farmer networks building endeavors. For those small farmers accessing HSSCs, working
through a farmer organization has become a necessary condition to remain in the channel.

Incentives offered by HSSCs in terms of higher profits and the opportunity of long-term

37

business relationships seem to offset or surpass the need for opportunistic behavior that

typically characterizes small farmer organizations.

2.5 Conclusions

The benefits accrued by small farmers who have successfully entered HSSCs
surpass those offarmers working in traditional markets. Issues of trust, coordination, and
resource endowments have surfaced as the major determinants of small farmers success.
While case studies are not used to provide inferential conclusions to other small farmers
around the world, this case study outlines the basic strategies and the difficult trade-offs
small farmer groups must undertake to emulate the behavior of private—sector companies.
After more than fifteen years in HSSCs, some farmers supplying foodservice have
reached a high level of competitiveness in a value chain highly sought after by larger
suppliers. This timeframe explains beyond doubt that with the key incentives from
markets, the necessary management techniques from group leaders and the minimum
small farmer asset endowments the issues of agency and transaction cost economics that
have constrained small farmers historically can be surmounted.

As an ex post facto analysis, the value chain assessment performed allowed the
understanding of how market trends shaped the different channels and allowed the
entrance of small farmers and their issues and solutions. While analyzing the data, the
findings had ramifications that touched more on social capital and cooperative theory
than it did on business and management theory. However, for reasons of study
delimitations, further expansion into those theories could not be completed. It is expected
that further research willexpand this attempt to turn value chain assessments into a more

rigorous research tool.

38

Finally, this value chain assessment meets its goals of providing the see-through
capability that was missing for this sector in preparation for other, more in depth studies
dealing with the success and failure factors affecting small farmers. The information
accumulated has been systematically organized in the following tables. Sources used in
each table are specified, and, in some cases only approximations were accomplished.
Nonetheless, these tables can be arguably the most complete compendium of available
information and analysis on a perishable product for Central America over the last ten

years.

39

Table 2.1: 2007-2008 Guatemala Lettuce Markets (Shares and Business Practices Gap Analysis)*

 

 

 

 

Traditional Channel HSSCs
Evaluation Factor Traditional Traditional Super- Food Total
GT ES markets Service
Total volume purchased (MT) 521 1.5 3424.7 5360.4 893.4 14890
Market share (% of total volume) 35 23 36 6 100
Suppliers associated permanently (#) 600 350 192 79 1221
Suppliers associated seasonally (#) 150 90 0 0 240
Wholesalers associated permanently (#) 2 I 4 1 l8 5 3 340
Wholesalers associated seasonally (#) 4O 25 0 0 65
Producers supplying more than one
channel (%) 5 2 45 5 57
Wholesalers supplying more than one
channel (%) 10 5 3 2 20
Product branding required None None Yes None
Product packaging (plastic wrap) No No Yes None
Food safety inspection required No No Yes Yes
Prices set per season No No No Yes
Prices set per week No No Yes No
Prices set per day (compared to spot
market) Yes Yes No No
Payment terms (days) Same day Same day 15-28 15-21

 

* Based on key interviews and cross validated through primary and secondary sources of
information

Table 2.2: Historical Data on lHSSCs

Year Outlets by Year
tradition
al

 

 

# of
Incep. store

 

Company markets 19 l9 19 200 200 # stores
year Elma were 89 94 99 4 9 urban-
used rural
last*** ratio
Wal-Mart (GT and 18
ES)* 1928 5 2003 42 87 9 209 241 3:1
La Torre (GT) 1950 2 2005 8 l l 12 26 45 7:1
McDonalds (C. A.) 1974 -- 1989 15 24 51 72 87 5:1
Pollo Campero (GT 13 18
and ES) 1971 -- 1989 67 2 9 221 227 6:1

 

*Owned by La Fragua until 1998; then under CARHCO until 2005, now owned 51% by Wal-Mart.
** Different store formants depend on neighborhood size, income level and purchasing culture.
*“Combined sources of information point out that at approximately 7% of total product sold in
supermarkets is still sourced from traditional markets.

40

Table 2.3: Chronological development of HSSCs in Central America

Market Channel Year-Event

 

Pollo
Decade La Torre McDonalds Campero
wamm (3" (GT) (GT and ES) (GT and
ES)
19205 1928: First Paiz store is
opened
19505 '1950: First store opens
19605 1960: First store in the rural 1964: Second store opens
area
19705 1974: 3rd and 4th store open 1974: First 1971:
restaurant open First
in Guatemala restaurant
open in
Guatemala
1978: 5th and 6th store open
19805 1983: 7th store with its own 1985: Global 1989:
bakery and deli standardized Food
food safety safety
assurance policies
adopted revised
19905 1995: La Fragua reaches 1993: First store opens in the
209 stores rural area (Escuintla)
1999: Royal Ahold buys 1999: Second store in the
33% of La F ragua stores in rural area (Retalhuleu) and
Guatemala and CSU stores two more stores in Guatemala
in Costa Rica under the City
name Central American
Retail Holding Company
(CARHCO)
1999: First attempt to
establish farm-based food
safety audits on key crops
20005 2005: Wal-Mart lntl. buys 2001: Merger with
51% of CARHCO Econosuper, changing name
to Unisuper totaling 26 stores
2006: List of Wal-Mart 2006: Supplier food safety 2007: 227
Stores: Guatemala 124; standards are requested restaurants
El Salvador 58; Honduras en Central
35; Nicaragua 33; Costa America
Rica 125
2009: Stores reach 520 2009: Total of 45 stores, 22 2009: 67 2009: 250
stores in Central America: La Torre format and 18 restaurants in restaurants
Guatemala 167; El Econosuper stores Guatemala and in Central
Salvador 82; Honduras 50; El Salvador America

Nicaragua 53; Costa Rica
168

 

Multiple sources: Retail and restaurant company websites, interviews, fact sheets, previous studies. See

references.

41

Table 2.4: Guatemalan Lettuce Suppliers Entry to Guatemala and El Salvador

 

 

 

 

 

 

 

 

Markets
Type of Firm Name Guatemala Markets year of El Salvador markets year of entry
entry
Tradi- Supers Food Spot Supers Food
tional Service sales Service
1. Private Royal 1998 1999 2000
company Antigua
2. Private La Carreta 1960’s Late Late Early
company with 1980’s 1980’s 1990’s (all
outgrower countries)
schemes
3. Farmer Labradores Early 2001 2005 1992 1992 to E1
Organizations Mayas 1 980’s Salvador;
2005 to
Costa Rica)
Horticultur 1993 I 997 1995
es Unidos
4. Wholesaler Lechugas 2002 2002
Marin
5. Wholesaler Disvegua Early 2006 1982 1985 I985
1980’s
Source: Interviews and Flores 2003.
Table 2.5: 2007-2008 Farm gate and Primary Gathering Site Cost Structure US$/Kg)*
Evaluation Factor Traditional Traditional Super- Food
GT ES markets Service
Production cost 0.75 0.75 0.80 0.82
Sorting/bagging/loading 0.05 0.05 0.02 0.05
Packaging/labeling 0.00 0.00 0.07 0.07
Product branding 0.00 0.00 0.05 0.05
Food safety inspection (required) 0.00 0.00 0.02 0.02
Transportation to delivery point 0.02 0.02 0.05 0.05
Total Cost 0.82 0.82 1.01 1.06
Average price paid at farmgate/delivery point 0.90 0.90 1.14 1.25
Average profit % 9.76 9.76 12.87 17.92

 

* Information collected per cuerda (l cuerda =1 166m2), pounds and local currency passed to
hectares, kilograms and dollars

42

Table 2.6: Perception of Small Farmer Assets GAP Analysis

 

 

Evaluation Factor Traditional Traditional Super- Food
GT ES markets Servrce

Social Assets
Reputation as an honest person low low medium high
Previous success in growing lettuce medium medium high high
Member of farmer organizations low n/a high high
Recommendations from other farmers medium low high high
Client of NCO-sponsored programs n/a n/a high low
Reputation with input suppliers n/a n/a high high
Previous loans and credits n/a n/a low low
Human assets
Literacy (knows how to read and write) n/a n/a high high
Age n/a n/a n/a n/a
Years of experience growing lettuce low low high high
Knowledge of quality and safety standards low low high high
Knowledge of improved varieties high high high high
Skills to grow at higher yields than average medium medium high high
Physical assets
Size of plots planted n/a high high high
Access to irrigation medium medium high high
Capacity to grow all year-round medium medium high high
Availability of other sources of income n/a n/a high high
Availability of warehouses and cold rooms low medium low medium
Availability of own transportation (truck) medium high high medium
Availability of greenhouse low low low low
Availability of small and large equipment low low low low
Environmental assets
Distance to the paved road high high medium low
Distance to the gathering locations high high high high
Distance to input suppliers n/a n/a high high
Distance to surface water (wells/rivers) medium medium low low
Distance to packing house high n/a high high
Moisture holding capacity of soil high high high high
Propensity to pests high high high high
Propensity to wind or hail damage high high high high
Availability of wells for irrigation high high high high

 

43

Table 2.7 : Guatemalan Lettuce* Production and Value 1989-2006 (000's MT and 000's $)_

 

 

 

Year Production (MT) Area (Ha) Value (000's USS) Average $/Kg
1989 6965 317 291 0.56
1990 6945 316 _ 300 0.61
1991 7976 363 283 0.61
1992 7856 357 314 0.60
1993 8054 366 333 0.61
1994 9056 412 399 0.73
1995 10266 467 429 0.75
1996 10981 499 489 0.65
1997 10284 467 496 0.74
1998 9297 423 435 0.97
1999 11056 503 633 1.02
2000 12463 567 759 1.03
2001 12276 558 764 1.05
2002 13098 595 839 1.09
2003 13783 627 1034 1.09
2004 14830 674 1072 1.10
2005 14010 637 1134 1.14
2006 14715 669 1104 1.28
* Iceberg lettuce is widely planted in Guatemala. Romaine varieties are also planted upon buyer's
request.

Source: Multiple sources (MAGA Central America export records, SIECA, exporting company
records).

44

Figure 2.1: The 2007-2008 Value Chain Map

 

Market
Channel
Share

Delivery
Point

 

 

 

 

 

 

 

 

 

Production!
Distribution
Market Share

 

 

 

Sorting
Packing

Wholesale
(Market share)

Production
(°/o farms)

Production
Origin

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   
 
 

Supermarkets Foodservice Traditional Traditional
GT/ES GT/ES GT ES
36% 6% 35% 23%

Teminal luMarke-tsul i'férini'riér'";
. . . . : Nearby EIMarketsl
Drstnbutron Processrng Markets iProductionli El E
Center Centers/Rest OTC"), l Areas liSalvadorE
L- .. J I- --
. ".3! """"
fi'\ f\ // 1' /—
.. - _ _ .. .. A- .1...J.'
”co/{nu 7%1 lWilson 2%: 2v{25v{3%"5v5 Isviifisvol

 
 

\,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-—-— —_-—

lSpot Markets: i Spot Markets i

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Central Highlands
64% Total Volume

 

 

HSSCs

45

 

Westem Highlands
36% Total Volume

 

 

 

TRADITIONAL MKTSi

.-_.s---- . .
Wholesaler : Packing I l__(_iga1ten1ala__g l.___ELS. lvad_o_r_ ..|
' I -—- ='-"—' _________
Pack'"g Plant r___P.'a"i.__' i Roadside/ l 1' Roadside L”
l l— [13203945211691 legrtiagenlnr
Large Dedicated
Dedicated Small Farmer Guatemalan Salvadoran
Wholesalers Organizations Wholesalers Wholesalers ¢_.:
(12)+12 = 24% 16% 40% 20% E
‘ ----- e 4 A A g
_ "‘-;T:_""—"'"'°"""""""'"""'T"' :
Medium farms! T" """" V! l
Wholesalers NT": Small farms r-——.i
(l2)°/o 880/0 :

 

l 000/0

1 000/0

100%

l 00%

l 000/0

Me 2.2: WaI-Mart Central America Store Formats and Number of Outlets

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Store Guatemala E1 Honduras Costa Nicaragua Total
Formats Salvador Rica

Supermarkets 30 32 7 25 7 101
Hipermarkets 6 2 l 6 1 5
Discount 1 15 48 27 126 46 372
stores

Warehouse 14 5 1 1 30
stores

Shopping 2 2
clubs

Total 167 82 50 168 53 520

 

Source: WaI-Mart Central America website.

46

 

Figure 2.3: Lettuce Tangible and Intangible Product Traits

Tangible
Traits

Intangible
Traits

Tangible
Traits

Intangible
Traits

 

 

 

 

  

 

 

 

Input Production Wholesaling Retailers
Sorting Foodservice
Packing Traditional
, _ _ Market _____
z . . . -Package 5
: -Cost of rnputs -lmproved-varreltres -Labe1ed : -Package d
: -Cost of servrces —Competrtrve prrce
. -Pre-cooked -Labeled
-Approved -Better handling
. . . -Same-day -Cooked
pestrcrdes -Longer shelf-life d l' T b'l'
-Seed variety -Traceability e rvery . . ' racea I "y
. _ _ —Traceabrlrty
quality _ . _
inspection Skillid growers . vSkilled handling Safety
. . -Qualrty certification . . guaranteed
-Safety Inspection . . -Trme1y delrvery . .
-Safety certification -Mrnrmum
-Improved . . -Brand support .
. -Relrable quality shelf-life
.stfisisncy. _________________________________________________________________________________
HSSCs 5
Traditional I
. ____________ i \5 Market ______
-Cost of inputs l -Marketable —Road picked
. ; appearance -Low cost
-Cost of servrces . -Poorly packaged . .
. . ; -Lower cost . -ered quality ,
-Tradrtronal i -Shorter shelf-life -Sh0rter shelf-life -No traceabili i
varieties ' -No traceability ty 1

 

 

E -No traceability

-Unknown quality of

water

-Unsafe practices
-Sun-exposure
l-Long

 

5 transportation time _

 

-Unprotected E
from hand '
touch

47

Figure 2.4: Michael Porter’s (1986) Value Chain Structure

 

Business management

 

Procurement

 

 

and

 

Research Processes

planning operations

and

 

 

 

Acquisition
marketing
and sales

 

. l . . .
Primary actrvrtres

48

 

Fi re 2.5: Research Site six communities in Chimaltenan 0 Guatemala

Comalapa
" e r
\ I I
: Santafiruz Balanya"
— v‘ '
I

V— ‘

I
\ \ r
Patzicia 9" -'
l
I
r Zaragoza

~\

 

 

 

 

BELIZE
MEXICO

 

 

GUATEMALA

*Guatemala HONDURAS
City

EL SALVADOR

 

 

Lettuce producing area

Source: Adapted from Guatemala National Institute of Geography (IGN 2004)

49

 

References

Adesina, A. A., and K. K. Djato. 1996. “Farm size, relative efficiency and agrarian
policy in Cote d' Ivore: profit function analysis of rice farms.” Agricultural
Economics 14. 93-102.

AGEXPORT (2008). Trade Data on Non-Traditional Exports. www.agexports.org. gt

Ankli, RE. 1992. Michael Porter’s Competitive Advantage in Business History.
University of Guelph.

Bagetoft, P., and H. B. Olesen. The Design of Production Contracts. Copenhagen
Business School Press. Hemdon, 2004.

Balsevich, F. 2006. Essays on producer’s participation in, access to, and response to the
changing nature of dynamic domestic markets in Nicaragua and Costa Rica. Doctoral
dissertation. Department of Agricultural Economics, Michigan State University.

Barham, B. L., M. R., Carter and W. Sigelko. 1994. Agro-export production and peasant
land access: Examining the dynamic between adoption and accumulation. Journal of
Development Economics 46: 85-107.

Barrett, CB. and T. Reardon. 2007. Making Markets Work for the Rural Poor. Faith and
Economics, October.

Beccerra, M., and A. K. Gupta. (1999). Trust within the organization: Integrating the trust
literature with agency theory and transaction costs economics. Administration Quarterly,
23, 177-203.

Berdegue, J .A., F. Balsevich, L. Flores, and T. Reardon. 2005. “Central American
supermarkets’ private standards of quality and safety in procurement of fresh fruits and
vegetables,” Food Policy, Vol 30 Issue 3, June 254-269.

Bromiley, R., and L. L. Cummings. 1995. Transactions costs in organizations with trust.
in R. Beis, B. Sheppard, and R. Lewicki (eds), Research on Negotiations in
organizations,5, 36-45.

Calvin, L., Foster, W., Solorzano, L., Mooney, J .D., Flores, L, & Barrios, V. (2002).
Response to a food safety problem in produce: A case study of a cyclosporiasis outbreak.
In B. Krissoff, M. Bohman, & J. Caswell (Eds), Global food trade and consumer demand
for quality (pp. 101-128). New York: Kluwer Academic Press.

Carletto, C., A. de Janvry and E. Sadoulet. 1999. “Sustainability in the Diffusion of

Innovations: Smallholder Nontraditional Agro-Exports in Guatemala. Economic
Development and Cultural Change. 47 (2): 345-369.

50

Cheung, SS. 1983. The contractual nature of the firm. Journal of Law and Economics
26(1):1-21.

Chow, I. H. 2008. How Trust Reduces Transaction Costs and Enhances Performance in
China’s Businesses. The Chinese University of Hong Kong.

Cook, M.L., and C. Iliopoulos. 2000. Ill-defined Property Rights in Collective Action: the
case of US Agricultural Cooperatives. Institutions, Contracts and Organizations:
Perspectives from New Institutional Economics. C. Menard, ed. Cheltenham: Edward
Elgar.

Cropp, R. 1989. Supporting organizations. In Cooperatives in agriculture, ed. D. Cobia,
98-105. New Jersey, USA: Prentice-Hall, Inc.

Downing, J ., M., Field and O. Kula. 2006. Globalization and the small firm: a value chain
approach to economic growth and poverty reduction. USAID Accelerated
Microenterprise Advancement Project. C02

Davis, SH. 1988. “Introduction: sowing the seeds of violence,” pp. 3-38 in Robert M.
Carmack (ed.), The Harvest of Violence: The Maya Indians and the Guatemalan Crisis.
Norman: University of Oklahoma Press.

Dries, L., and J .F.M. Swinnen. 2004. Foreign Direct Investment, Vertical Integration and
Local Suppliers: Evidence from the Polish Dairy Sector. World Development, 32(9), pp.
1525-1544

E1 Periodico. (2006). Las manos detras de una mordida (McDonald’s Central America source
of raw material). www.clperiodico.com.gt/e5/20060903/actualidad/31558

Eisenhardt, K. M. (1989). Agency theory: An assessment and review. Academy of
Management Review, 14,57-74.

Farina, E. and T. Reardon. 2000. “Agrifood grades and standards in the extended
MERCOSUR: conditioners and effects in the agrifood system,” American Journal of
Agricultural Economics, 82(5), December: 1170-76.

Feller, A. 2006. Value Chain versus Supply Chains. BPTrends.

Granados, J .C., 2004. “Analis del Comercio Regional y Extragional de Frutas y
Hortalizas en Centro America”. PF ID-F&V, Michigan State University. Report not
published.

Green, LB. (1989) ‘Consensus and Coercion: Primary Health Care and. the Guatemalan
State’, Medical Anthropology Quarterly 3(3): 246—57.

51

Gulati, A., Minot, N. Delgado, C. and S. Bora, 2006. Growth in high-value agriculture in
Asia and the emergence of vertical links with farmers. In: Swinnen, J .F.M (ed) Global
Supply Chains, Standards and the Poor, CABI publishing.

Hamilton, S. G, Fischer, E. F. 2005. Maya Farmers and Export Agriculture in Highland

Guatemala: Implications for Development and Labor. Latin American Perspectives,
2005; 32; 33 Latin

Hamilton, S. G. and Fischer, E.F. 2003. Nontraditional agricultural exports in highland
Guatemala: understandings of risk and perceptions of change. Latin American Research
Review 38 (3): 82—110.

Hart, 0. 1995. Firms, contracts, and financial structure. Oxford, UK: Oxford University
Press.

Hill, C. W. L. (1995). National institutional structures, transaction cost economizing and
competitive advantage. Organization . Science, 6(1), 119-131.

Hernandez, R., T. Reardon, and J .A. Berdegue. 2007. “Supermarkets, Wholesalers, and
Tomato Growers in Guatemala,” Agricultural Economics, 36(3), May.

Hu, D., Reardon, T., Rozelle, S., Timmer, P.,Wang, H. 2004. “The emergence of
supermarkets with Chinese characteristics: Challenges and opportunities for China’s
agricultural development.” Development Policy Review 22 (4), 557—586.

James J .W., Glenn H. Sullivan, and Guillermo E. Sanchez 2000 “Future market
development issues impacting Central America’s nontraditional agricultural export

sector: Guatemala case study.” American Journal of Agricultural Economics. 82: 1 177—
1183.

Jano, P., M. Lundy, F. Balsevich, J. Berdegue, and T. Reardon. 2004. “Farmers’
organizations and supermarkets in Central America”, Report. Rimisp—Latin American
Center for Rural Development, Santiago, Chile.

Jones, E. 2004. Wealth-based trust and the development of collective action. World
Development 32 (4): 691—71 1.

Kaplinsky, R. and Morris M. 2000. A Handbook for Value Chain Research.

Key, N., and D. Runsten. 1999.”Contract farming, smallholders, and rural development
in Latin America: The organization of agroprocessing firms and scale of

outgrower production.” World Development 27: 381-401.

Kirsten, J ., and Sartorius, K. 2002. Linking agribusiness and small-scale farmers in
developing countries: 15 there a new role for contract farming? Development Southern

Africa 19(4):503-529.

52

Kula, Olaf; J. Downing and M. Field. 2007. Globalization and the small firm: An
industry value chain approach to economic growth and poverty reduction. Micro Report
42. USAID.

Lamb, J. 1991. Final Report on Proexag Non-Traditional Agricultural Export Support
Project. Chemonics International on line records.

MAGA 2008. Ventanilla Unica para Ias Exportaciones. Non-published trade data.
Magretta, J. 2002. Business Models Matter. Harvard Business Review.

Martinetz, S. 2005. “Vertical coordination of marketing systems: Lessons Ieamed from
the poultry, egg and pork industries.” United States Department of Agriculture.
ERS Agricultural Economic Report No. 708. May 2005.

Nooteboom, B., H. Berger and N. G. Noorderhaven. 1997. Effects of trust and
governance on relational risk. Academy of Management Journal, 40,308-338.

OECD. 2008. Enhancing the Role of SMEs in Global Value Chains ISBN
9789264050952.

Okello, J .J ., and SM. Swinton. 2007. “Compliance with International Food Safety
Standards in Kenya’s Green Bean Industry: Comparison of a Small and a Large Scale
F arm Producing for Export.” Review of Agricultural Economics 29(2): 269-285.

Orellana, E. 2003. USDA FAS Report on Foodservice El Salvador (check reference)

Pomareda, C. 2001. Small Farmers and their Participation in Central-American
Agricultural Exports. UNCTAD Regional Workshop Proceedings.

Porter, M. 1985. Competitive Advantage. Creating and Sustaining Superior Performance.
New York.

Reardon, T. and L. Flores. 2006b. “Viewpoint: ‘Customized Competitiveness’ Strategies
for Horticultural Exporters: Central America Focus with Lessons from and for other
Regions,” Food Policy, 31(6).

SIECA Trade Data. 2004, 2007. Integrated Commerce lnforrnation System for Central
America. www. sieca. org. gt

Smallbone, D., and F. Welter. 2001. The distinctiveness of entrepreneurship in transition
economies. SmallBusiness Economics, 16(4), 249.

Sykuta, ME. and Cook, ML 2001. A New Institutional Economics Approach to

Contracts and Cooperatives. American Journal of Agricultural Economics, 83, N°5:
1273-1279.

53

Von Braun, J. Hotchkiss, D., and Immink, M. 1989. “Nontraditional Export Crops in
Guatemala: Effects on Production, Income, and Nutrition,” Research Report no. 73
(International Food Policy Research Institute, Washington, DC, 1989).

Williams, Robert G. 1986. Export Agriculture and the Crisis in Central America. Chapel
Hill, NC., USA. University of North Carolina Press.

Williamson, 0. The Economic Institutions of Capitalism: Firms, Markets and
Relational Contracting. New York: Free Press, 1985.

Willis, K. 1990, “In-depth interviews”, in Birn, R., Hague, P., Vangelder, P. (Eds),A
Handbook of Market Research Techniques, Kogan Page, London, pp.249-60.

www.aboutecoli.com 2007. URL: http://www.about-
ecoli.com/ecoli_outbreaks/view/jack-in-the-box-e-coli-outbreak

Zaheer, A., McEvily, B., Perrone, V. (1998). Does trust matter? Exploring the effects of
interorganizational interpersonal trust on performance. Organization Studies, 9, 141-159.

54

CHAPTER 3
A CASE STUDY OF SMALL FARMER ORGANIZATIONAL STRATEGIES TO
INCREASE MARKET ACCESS PERFORMANCE

3.1 Introduction

The literature from non-traditional agricultural exports (NTAX) adoption and the
increasing market share of supermarkets in Central America suggests that small farmers
affiliated with organizations are more likely to leapfrog from traditional agriculture to
more profitable market venues as they improve their access to information, inputs, credit
and technical assistance collectively (von Braun et al., 1989; Carletto etal., 1999;
Hamilton and Fischer 2005; Berdegue et al., 2005). However, in the case of small tomato
farmers accessing supermarkets in Guatemala, Hernandez et al., (2007) described that
belonging to a farmer organization does not necessarily indicate a direct improvement in
market access performance. Since some organizations are not designed for that purpose.
Yet, as described by Jano and Mainville (2006), even when some organizations are
formed specifically to promote market access for all their members, they often fail to
reach this goal. The literature on cooperatives and farmer organizations points out how
these challenges are typically related to limited access to information, lack of
management skills, inadequate cash flow, insufficient income generation and poorly
structured monitoring and enforcement of compliance with market requirements
consistently (Ortmann and King 2006).

For decades, several cooperatives and farmer entrepreneurial groups have
organized around market expansion goals with mixed results (Jano and Mainville 2003).
However, little in the recent literature has focused on effective organizational strategies

that have succeeded in accessing modern markets such as HSSCs successfully. Most of

55

the existing literature on associations attempting to access high standard supply chains
(HSSCs) such as supermarkets and foodservice companies depict a challenging rather
than promising story for small farmers (J ano and Mainville 2006; Okello and Swinton
2007).

A question in the debate would be what could we learn from a farmer
organization that is highly selective on itschoice of members to ensure success in
accessing target markets? This is the case of Labradores Mayas (LM) which has accessed
HSSC since 2001 and has increased its market share in this demanding market channel
for nearly ten years. LM is not a conventional farmer organization where every member
has vote and voice. The evolution of LM’s organizational strategies could be categorized
as a hybrid mixof a cooperative and a private production and wholesaling company that
has gone through important trade-offs in the process. In assessing the success of LM this

essay is focusedson two central research questions:

0 How has LM arrived at this level of market access performance in HSSCs?

o How can the LM organization model be characterized for other farmer

organizations to pursue access to HSSCs?

The study was carried out in six villages the Guatemalan Highlands where LM
members are located. This area has also become the major catchment area for HSSCs in
recent years due to its agro-climatic conditions, proximity to the panamerican paved
highway leading to Guatemala City, and the years of experience in the area to produce

cool-weather vegetable crops.

56

LM’s nearly ten years of continued market access represents a “critical case” in
testing the theory on small farmers exclusion from HSSC’s outlined in recent literature.
By studying this organization’s pathway to market access this paper contributes to the
literature on small farmers’ access to modern markets on two major points. First, it
provides further understanding on farmer organizational strategies to leapfrog from local
traditional markets to more demanding HSSCs. And second, it builds on the theory of the
necessary farmer asset levels to access and compete in profitable markets where larger
farmers are typically the most suited to compete (Maertens and Swinnen 2007).

This chapter is organized in three additional sections. Section 3.2 provides a
historical background on Labradores Mayas. Section 3.3 describes the theoretical
framework on farmer organizations accessing modern markets. Section 3.4 discusses the
results from interviews outlining market evolution and strategies adopted by LM, while

Section 3.5 offers conclusions.

3.2 Historical review of Labradores Mayas

Founded in 1993 by three family members, LM has expanded its membership
base consistently over the past sixteen years. In a first attempt to study this group, Flores
(2002) reports that by 2001 the organization had sixty-six members located in the
municipalities of Patzicia, Santa Maria Balanya and Tecpan Guatemala, with occasional
land rentals in nearby communities around specific months of the year. By 2008 it was
confirmed that the membership base had reached 136 active farmers. According to their
leaders, LM planned to reach 200 members by the end of 2010 even though the global

economic slowdown has affectedCentral American countries.

57

In 16 years of existence LM members have leapfrogged from road-side sales to
more strict market channels in El Salvador and Guatemala. By 1998, LM had entered
their first formal contract with a large wholesaler supplying Salvadoran supermarkets.
Later in 2001, its first direct supply contract with supermarkets marked a new era of
modernization to adapt to higher quality and safety practices. To remain in HSSCs meant
the adoption of more coordinated management practices to ensure that members followed
new production protocols ensuring uniform quality. Table 3.1 shows the evolution of LM

through different markets.

The case of LM is unique among cases of organized farmer groups. This is
because LM has adapted over the years from less to more complex demand conditions
established by different buyers in complying with stringent quality, safety and supply
chain (logistics) standards. Empirical evidence from South America and Africa shows
that farmers with larger asset endowments have been better prepared to comply with such
standards (Reardon and Farina 2001; Okello 2006). Also, recent research on the
emergence of supermarkets in Latin America shows there is a trend in supermarkets to
stock up on large producers with more access to capital. Larger farmers have historically
complied better with demanded volumes, quality, logistics and specific standards, thus

reducing transaction costs (Mainville and Zylbersztajn 2005).

The case of LM illustrates that there are proven organizational strategies small
farmer groups can apply in accessing dynamic markets. Although it is a long process
rather than a rapid adaptation to change, those organizational strategies have been

possible and practiced over time. According to interviews, some of the incentives in

58

adopting new HSSCs penetration strategies are typically higher profits and the ability to
enter into formal supply contracts and convenient fixed prices (above traditional market
places). Interviews pointed out that twenty years ago, most members of LM lived in
poverty with annual incomes of $800-$1 500 growing corn and beans in small plots of
land. Now LM members are supplying lettuce, carrots, celery and other vegetables under
stringent quality and safety standards to Wal-Mart Central America and other buyers. The
annual income levels of LM members range from $4500-25,000 depending on the
extension produced, a success story in income expansion that took 16 years in the making

and continues to expand.

How did LM reach this level of organization and professionalism in producing,
sorting, packing and delivering their products? This question was addressed through
interviews with member farmers that the LM story has similarities with hundreds of
farmers in the highlands who have accessed new income generation opportunities
through the production of high-value crops for export (NTAX). The NTAX trend started
with the launching of the US Caribbean Basin Initiative (CB1) in the early 1980's, which
provided tariff-free import status to all non-traditional fruits and vegetables from the
Central American Region (von Braun et al., 1989). The "non-traditional" adjective was
created to differentiate these products from crops traditionally grown in the region such
as bananas, sugar, cotton and coffee. Following the enactment of the CBI, several US-
financed development projects provided research on the adaptability and yielding
research of different non-traditional crop varieties targeted to the US and European
markets (Lamb 1991). Different key informants coincided that farmers participating in

export markets in the central highlands improved their incomes at staggering annual rates

59

between 1991 and 1996 (key informant interview 2008). James et al., (2000) report that

this growth was 16% annually between 1983 and 1997.

LM and NTAX producers both benefited from the new agricultural production
knowledge and practices introduced in the 19805 and 19905 by development projects (key
informant interview 2008). As one of the NGO-representatives interviewed put it “The
NTAX trend of the 19805 and 19905 was like a mini green-revolution in the Guatemalan
highlands that brought knowledge on improved varieties, integrated pest management
and modern fertilization techniques. This also improved the capacity of small farmers to
supply the local and regional markets with improved varieties slowly creating a demand
for better agricultural inputs for crops typically consumed in urban markets in Guatemala

and El Salvador.”

3.3 Theoretical framework

Studying LM’s strategies to compete involves comparing and contrasting what
LM has done with what other single agricultural business and collective agricultural
organizations have done with similar or different market access performance. This study
draws on microeconomics (at the level of households and the organization) and collective
action theory concerning how smallholder organizations gain access to HSSCs through
entrepreneurial organizational schemes, (Cook and Plunkett 2006). The emergence of
new forms of entrepreneurial group organization has been documented since the 19905 in
North America, Oceania and Europe (Cook and Plunkett 2006). Their rise is not only in
response to outside pressures to compete associated to technology and a globalized

market economy, but also due to the internal needs to achieve better coordination and

60

redirecting their strategies to changing rules on the demand side (Faulkner and De Rond
2000; Comforth 2004; Cook and Plunkett 2006; Barham and Chitemi 2008; Muradian

and Mangnus 2008).

Recently, Chaddad and Cook (2004) characterized new forms of emerging
cooperative organizational models on the basis of property structures. While most of the
examples recently quoted in the literature apply to cooperative forms in developed
country scenarios, Muradian and Mangnus (2008) points several challenges with the issue
of small and poor agricultural producers organizing to compete. Particularly, Murandian
and Mangnus outline the issues of horizon and support to unprofitable organizational
structures, tow often-quoted concerns in the development and cooperative literature
(FAO 2006). Supporting new forms of entrepreneurial organization is not new for
international development efforts. However, the recent troubled past of even new
generation cooperatives structures motivated other forms of organizational management
in developed and developing countries. The general aim has been to avoid the classic
dilemmas of the classic cooperative organizational and management model (Ottmann and

King 2006).

Important insights to this research are obtained from a large-scale organizational
development support in Chile during the 19905 supported targeted farmer groups to
become entrepreneurial organizations to increase collective bargaining power vis-c‘r-vis
value chain links (Berdegue 2001). Through this effort farmers were organized into
Associative Peasant Business Firms (APBF) where small-scale farmers work as owners

and thus control the organization’s decision-making process from production to market

61

access activities. Berdegue’s findings point out several positive outcomes, particularly
on improving APBFs access to HSSCs where transaction costs and value-addition
technology is needed. However, it is also outlined that not all APBFs have ended in
success stories. For this initiative to be more effective, the author recommended major
improvements to this program in the order of (a) more effective approach to improve the
performance of APBF members as independent farmer firms in a market economy; (b)
promoting a structure of self-sustainability (away from government subsidies) as true
business firm, and; (c) ameliorating the APBF strategy to become a more institutionally

robust social platform for collective action (Berdegue 2001).

The internal issues cooperatives have faced in recent years in terms of cash flow
and income generation, agency, increasing cost of energy and other resources have been
quoted as major reasons motivating the rise of new forms of organization (Gray and
Kraenzle, Faulker and De Rond 2000). However, external issues outside the scope of the
organization per se have also shocked the speed at which organized farmer groups are
able to compete. Several authors document the increased quality, safety, labor and
environmental standards that affect the way wholesalers and retailers work with suppliers
(Calvin et al., 1999; Berdegue 2001; Berdegue et al., 2005 and 2006; Downing and Kula

2006; Okello 2006).

These changes have affected farmers regardless of production scale, geographic
location and capacity to access markets, which demands a new approach to
competitiveness. For farmers around the world, particularly in developed countries, this

means moving beyond classic organizational models into more streamlined management

62

structures that facilitate the decision making processes to compete. For small and poor
farmers in developing countries with the need for public- or NGO-led development
programs to assist them in organizing to compete, the process poses significant
challenges. An important reflection on this subject is offered by Muradian and Mangnus

(2008) as follows:

“The development sector has traditionally stressed the importance of farmer
organizations primarily as tools of political empowerment, advocacy and
representation. This was a logical stand at a time when the state played a more
important role in steering rural development, setting prices of agricultural
products, allocating resources and regulating commercialization. It may yet be
misleading nowadays in countries with liberalized agricultural markets and in a
context of the growing importance of global agri-food chains. As for their market
function, development practitioners have — in part due to their suspicion of
(especially large) private commercial agents — promoted cooperatives as
countervailing power to existing market forces, which in many developing
countries reflect historically very unequal class relations. Above all, cooperatives
are conceived by the development sector as agents of social change. They are
expected to raise the voice of their constituency against prevalent unjust social
and economic relations, and to be an engine of local development. The
development sector therefore tends to stress the notion of social entrepreneurship

in cooperative development.”

63

J uxtaposed to the idea of farmer organizational groups being tools of political
empowerment, advocacy and representation are the notions of entrepreneurship and
competitiveness that other industries under liberalized trade have been exposed to for
decades. This excerpt also brings to attention that while cooperatives remained the
showcases of poverty reduction strategies, the long-term support through direct or
indirect subsidies from technical assistance to procurement of inputs does not create a
self-sustainable business sector as seen in the literature from Chile and Latin America
(Berdegue 2001; Pomareda 2001; Jano et al., 2004; Helling et al., 2006). In the face of
this dilemma, Hart and Moore (1990) reflected almost twenty years ago that under
liberalized trade environments agricultural companies and organizations were already
under constant pressure to compete and Ieamed to survive while others disappeared. This
is the reality developing countries have also been facing over the past two decades as
pointed in the evidence from Brazil, Chile and Guatemala (Farina and Reardon 2000;
Mainville and Peterson 2005; Mainville and Zylbersztajn 2005; Jano and Mainville

2006).

In studying how LM succeeded it is also important to draw on the literature on
business case studies. In business case studies the statistically-outlying cases matter as
others can learn about how decisions were made and strategies tried that led to a positive
or negative outcome (Hart and Moore 2008). Most business case studies proceed with a
logical sequence of evaluations to cover at least seven areas: (1) an analysis of the
company’s history, development and growth; (2) identifying the company's internal
strengths and weaknesses; (3) analyzing the external, enabling environment; (4)

performing a strengths, weakness, opportunities and threats (SWOT) analysis; (5)

64

analyzing corporate-level strategy; (6) analyzing business-level strategy; and (7)
analyzing structure and control systems to evaluate whether that structure is the
appropriate one for the company (Porter 1991). In this study, LM is approached with a
similar set of analytical tools to understand how they have arrived to over ten continuous

years of competitive performance in HSSCs.

3.4 Case study methods

The case study approach has been used to answer the how and why questions in
empirical, qualitative research. Yin (1989) defined a case study as an empirical inquiry
that investigates a contemporary phenomenon within its real-life context when the
boundaries between phenomenon and context are not clearly evident and in which
multiple sources of evidence are used. Guba and Lincoln (1989, 1994) explain that
qualitative research aims at gathering an in depth understanding of human behavior and
reasons governing it.

In answering how LM arrived to its level of market access performance and how
its organizational model can be characterized for other organizations to learn from their
experience, five areas of inquiry were pursued: (1) Understanding how LM managers
consider themselves different from other organizations. (2) Understanding the context of
threats and opportunities within which LM evolved. (3) Focusing on “process” rather
than “outcome.” This does not mean focus on outcomes is not important, but rather it
means that understanding the process that leads to outcomes is a major strength that
experimental and survey research seldom times do not address in depth. (4) The

possibility of developing causal explanations to better categorize LM’s organizational

65

model and management structure in a way that other organizations can explore similar

models realizing the trade-offs.

Following this reasoning, to engage in this empirical inquiry, findings in the
preceding value chain study provided valuable insights to develop a focused, structured
interview guide to last from 60-90 minutes with the leaders and founders of these
organizations. Some of the questions had an open-ended format, but the questionnaire

was framed to collect specific data needed for a case study protocol.

Data gathered through these interviews were recorded in a field journal and
analyzed to identify plausible explanations to LM decision making process over the
years. Data analysis and procedures were detailed in the case study protocol to serve as a
logbook of different activities from budget notes to report outlines following several

authors’ recommendations (Yin 1989; Maxwell 1996; and Rubin and Rubin 2005).

3.5 Results

3.5.3 Access to HSSCs as an organizational shaping factor

Founded in 1993 by 16 members, LM started selling to the traditional wholesale
market in Guatemala City. The initiative started with lettuce and carrots, the main
products of the area and with which LM members had more experience. As in other
wholesale markets, there was a high degree of speculation on prices. Based on the
experience of selling their product as an organization and moving away from their
traditional markets, LM farmers acknowledged that their ability to produce high quality
with other experienced farmers in the community could satisfy more demanding markets.

Following suggestions from key informants, the next step was to supply wholesalers

66

supplying supermarkets in El Salvador. The company Disvegua was the first experience
for LM requiring production practices with higher standards. Disvegua required that the
product should be placed in plastic boxes and fill weight, size and packaging
requirements. Table 3.1 explains the chronology of market entry of Labradores Mayas
while Figure 3.1 describes the differences between supplying to traditional markets and

HSSCs.

LM’s primary objective was to design a working model that would allow a
flexible management decision making. LM also wanted to recruit members recognized
for the agronomic management of specific crops, ability to follow instructions and their
commitment and understanding of the association’s long-term plans of the association. Its
general objectives were to urgently reach a competitive level in the local market
eliminating the need for intermediaries. In addition, leaving the management in the hands
of a small three-member team provided swift decision making to recruit farmers with a
minimum level of land, access to water to produce several cycles per year, and who
agreed to use inputs delivered by LM under strict production management schedules. For
instance, some pesticides were measured and delivered every morning to the members to
be applied in the field the same day. Such controls, as if a large farm owner divides
measures pesticides in the required quantities for his workers to apply, gave LM the same
capacity to comply with pesticide residue standards in demanding markets. Cooperation
among its members to enter better paying markets led to the expansion of the collection

and packaging hangar.

67

The conquest of more sophisticated markets began in El Salvador, where a LM
supplied chain Super Selectos supermarkets for over two years. Super Selectos had fifty-
six outlets and a history of fifty years of existence in El Salvador. The chain recognized
the quality of work and reputation with Disvegua standards in the management of weight,
color and other quality characteristics (Marin 2003). There were three major challenges
to stay in this market channel through Disvegua. First, the toughest test for LM in the
direct supply to Super Selectos was the level of logistics required to deliver their products
timely and in the quantities requested. Second, the issue of credit was also a difficult step
to overcome since Super Selectos worked on a basis of 30 days to pay. However, the
incentive of prices between 15 and 30% above the wholesale markets gave LM the
propelling force to deliver bi-weekly shipments. The third challenge was the “shrinkage”
factor. This term is applied to the shortages in product after delivery due to quality
problems (blemishes, loss of weight and acceptable appearance). Super Selectos imposed
a fee of 5% automatic discount on the total value of the transaction which suppliers
accept often without verifying the level of losses. Despite this discount, LM has realized
that the return on direct sales to supermarkets, even if through Disvegua, was still
superior to traditional wholesale markets. Figure 3.1 summarizes the number of activities

necessary to comply at the wholesale and retail level in HSSCs.

3.5.2 LM strategies to access HSSCs directly

Modernization in the procurement of fruits and vegetables has introduced
important elements of organization and institutional change in the retail sector as well as
in the food service sector over the course of the last fifteen years (Reardon et al., 2003;

Berdegue et al., 2005; Orellana 2003). As it has been described in Chapter 2, these

68

changes have given origin to different market channels for lettuce that moved away from
fragmented supply chains in traditional markets, becoming full value chains in HSSCs. In
the same time frame, multiple seasons since the mid and late 19905 indicate that the
consolidation of commercial relationships between LM and its market channels was also
a process of several years. Therefore, deep changes in structure and conduct of markets
have been necessary to facilitate the participation of LM, and these changes took time.
For instance, at the small farmer group level, repeated transactions took place leading to
the assumption that association members switched from opportunistic behavior to higher
levels of cooperation.

As HSSCs evolved from no inspections to a gradual increase in levels of quality
and safety standards required, LM was able to demonstrate its capacity to adapt by
fulfilling all the required conditions. It is not typical for small farmers to respond quickly
to standards required by HSSCs, particularly in a product such as lettuce. Compared to
less perishable commodities such as onions and carrots which can be thoroughly washed
or cooked before consumed to take care of microbial hazards, lettuce is a more difficult
product to deal with. The level of attention required in lettuce increases transaction costs

directly linked to tangible and intangible quality, safety and delivery requirements.

As these changes took place, LM management has also evolved over time to carry
out more functions downstream in the value chain delivering a final, packaged, safety-
inspected product to the retailer. To operate in a similar fashion as the competing, larger
companies, LM has adopted a management structure that eliminates the need to consult
with all of their suppliers on the direction of the business strategies. This may not be

positive in comparison to other organizational models such as new generation

69

cooperatives; however, LM believes their strength on reacting swiftly to necessary action
lies on this management model. For instance, according to one of the managers,
responding to market opportunities, terminating supply relationships with non-complying
members and negotiating larger market shares and quality premiums would be largely
handicapped if LM had a consultation process outside the management team for all key
decisions. It was the opinion of one of their long-time leaders that a slow decision making
process has led other groups to failure and that is where LM has shown a management
advantage. In recent years, few members have complained about this system due to the
continuous positive experiences season after season. Positive experiences for LM
members mean profit levels that other marketing alternatives such as the traditional
market channels have not matched. LM clients, particularly supermarkets, are satisfied
with this management style and have continuously rewarded LM with opportunities to

grow their weekly supply volumes (key informant interview 2008).

Based on LM responses to questions concerning how the LM organization model
can be characterized for other farmer organizations to pursue a similar development path,
four major themes are explained. These are organizational strategies that have resulted in
a positive result, but have also implied significant tradeoffs. For instance, LM had to
adopt to enter HSSCs and remain active despite the systematic elimination of some of its
underachieving members. Figure 3.2 represents these four actions as a

challenge/action/result pathway to sustained market entry.

70

Benchmarking group performance against larger groups and companies

LM members expressed that they faced difficulty realizing their competitive
strengths and weaknesses without comparing their operations and performance to similar
or better management models. Specialized wholesalers are typically medium- or large-
scale companies that have succeeded in agricultural value chains in Guatemala,
particularly in the NTAX industry. Successful small farmers have followed the path of
older entrepreneurial groups and companies to become more conscious about their
potential shortcomings. This has been true in the case of opting for more expensive, but
also more effective, technology, acquiring loans, and managing members’ failures. As a
result, their cost structures were believed to be very similar. Constant benchmarking of
what small farmer groups do against better—performing, larger groups and private
companies has been an eye opener for successful groups supplying lettuce. LM’s
decisions include what truck to buy and what upgrades to make in their operations on a

constant basis.

Increasing efficiency through expansion of networks

LM leaders expressed the importance of accessing information in the value chain
to improve their know-how from basic tasks to major decisions such as changing lettuce
varieties. Most of the interviewees remembered how it was through connections in their
immediate network with other businesses that they first heard of opportunities that
became important business leads. It is also through network members that information
flows effectively concerning reputable farmers that could become future members. The

upshot of building networks through mutually beneficial relationships is the capability to

71

keep abreast with key aspects affecting cost, quality, volume, and logistics in the value
chain. Networks upstream permit access to better input prices and technology
improvements, while networks downstream play an important role in passing down

growth opportunities and managing risk and problems before these are out of control.

Controlling origin, distribution, and use of key production inputs

When LM was faced with tougher grades and standards such as food safety
compliance, managing pesticides and other inputs became a necessary—additional"
management task. LM leaders outlined that ensuring compliance with expected quality
standards is critical to remaining successful. These tasks involved contracting with input
suppliers, supervising delivery of inputs, distributing critical inputs such as pre-measured
pesticides and monitoring members to identify deviations from the production and
handling guidelines. In the past, ignoring the importance of these tasks has led to lack of

product trace-back capability and problem-solving capacity.

Working with growers who minimize costs and increase group performance

Working to keep up with increasing standards, managing tighter schedules, and
complying with tangible and intangible product traits are tasks typically burdensome for
most small farmers wishing to stay in the value chain. According to LM leaders
interviewed, some members have failed to stay in the group because they do not follow
instructions as requested, particularly concerning pesticide application. In either case,
Successful farmer groups face the tough decision of losing members who voluntarily give

up and decide to supply less-demanding market channels, while other members have to

72

be excluded for not followingthe required protocols. When asked what is required of
members in terms of minimum assets, the responses corresponded greatly to the Table
3.3. LM interviewees emphasized the capacity to produce all-year round as a major
characteristic of their members. Availability of water for irrigation and specific locations
where hail damage is less likely were two of the most desired characteristics.

While this has been a necessary condition to stay competitive, it has in many
ways damaged the reputation of the organization in the community from being no
different than private, large wholesalers protecting their brand by working with a specific
pool of convenient farmers to their business. The situation is such that reputation over the
years has been the key to expand LM market share within their targeted market channels.
Losing or maintaining this reputation is a situation that depends on all and each of the

members.

3.6 Conclusions

One of the key challenges for the development sector in supporting small farmer
organizations is how to transfer an entrepreneurial spirit in markets with changing and
demanding requirements to compete. The entrepreneurial capacity of LM has been
remarkable over the past ten years and is worthy of outlining for others to learn. The
decision making process LM has engaged into for the past ten years has been oriented
towards ensuring a strong competitiveness position with HSSCs. However, along the
way, tradeoffs had to be taken not only at the level of individual relationships with its

members, but also on the mixed image they have built for themselves in the community.

Escaping the fate of agricultural c00peratives in developing countries that lost

control of their future, LM stands as an example of a grower entrepreneurial group that

73

has created remarkable economic benefit for its members along the years. In belonging to
the LM group, members do not value any other benefit above the opportunity of higher
profits, something previous participation in cooperative-like organizations could not
provide. The downside of their success is, as pointed out in the literature, the exclusion of
small farmers that do not have the necessary experience and the minimum set of assets to

produce according to standards.

Four key organizational strategies describe the LM model: (1) Working with growers that
minimize costs and increase group performance. (2) Controlling origin, distribution, and
use of key production inputs. (3) Increasing efficiency through expansion of networks.
And (4) benchmarking group performance against larger groups and companies.
Following those strategies has built a promising future for its members and they are
poised to continue satisfying demanding clients such as Wal-Mart and expanding into the
Central America Region. These four strategies are based on emulating how private sector
competitors work and on avoiding some of the issues faced by farmer groups that failed.
Through this detailed study the evolution, organization, management and modus
operandi of LM is exposed outlining the critical points in their path to success. A few
years from now future research should focus on defining whether this form of farmer
organization is sustainable and replicable to areas within and outside the conditions of the

Guatemalan Highlands.

74

Table 3.1: Guatemalan Lettuce Suppliers Entry to Guatemala and El Salvador
Markets

 

 

 

 

Type of Firm Name GT Marketsyear of entry ES markets year of entry
Tradition Supers Food Spot Supers Food
a1 Service sales Service
1. Private Royal 1998 1999 2000
company Antigua
2. Private La Carreta 1960’s Late Late Early
company with 1980’s 1980’ 1990’s (all
outgrower 5 countries)
schemes
3. Farmer Labradores Early 2001 2005 1992 1992 to E1
Organizations Mayas 1980’s Salvador;
2005 to
Costa Rica)
Horticultur 1993 1997 1995
es Unidos
4. Wholesaler Lechugas 2002 2002
Marin
5. Wholesaler Disvegua Early 2006 1982 1985 1985
1980’s

 

Source: Interviews and Flores 2003.

75

Table 3.2: Percgption of Small Farmer Assets Gap Analysis

 

 

Evaluation Factor Traditional Traditional Super- Food
GT ES markets Servrce

Social Assets
Reputation as an honest person low low medium high
Previous success in growing lettuce medium medium high high
Member of farmer organizations low n/a high high
Recommendations from other farmers medium low high high
Client of NGO-sponsored programs n/a n/a high low
Reputation with input suppliers n/a n/a high high
Previous loans and credits n/a n/a low low
Human assets
Literacy (knows how to read and write) n/a n/a high high
Age n/a n/a n/a n/a
Years of experience growing lettuce low low high high
Knowledge of quality and safety standards low low high high
Knowledge of improved varieties high high high high
Skills to grow at higher yields than average medium medium high high
Physical assets
Size of plots planted n/a high high high
Access to irrigation medium medium high high
Capacity to grow all year-round medium medium high high
Availability of other sources of income n/a n/a high high
Availability of warehouses and cold rooms low medium low medium
Availability of own transportation (truck) medium high high medium
Availability of greenhouse low low low low
Availability of small and large equipment low low low low
Environmental assets
Distance to the paved road high high medium low
Distance to the gathering locations high high high high
Distance to input suppliers n/a n/a high high
Distance to surface water (wells/rivers) medium medium low low
Distance to packing house high n/a high high
Moisture holding capacity of soil high high high high
Propensity to pests high high high high
Propensity to wind or hail damage high high high high
Availability of wells for irrigation high high high high

 

76

Figure 3.1: Lettuce Tangible and Intangible Product Traits

 

In ut Production Wholesalin Retailers
P g
Sorting F oodservice
Packing Traditional
Market ......
E -Cost of inputs 5 -lmproved varieties 5 £335: E -Packaged
Tangible 5 -Cost of servrces l -Competrtrve.prrce 5 -Pre-cooked 5 -Labeled
. : -Approved : -Better handling : :
Traits - - - - . . -Same-day . -Cooked
: pestrcrdes : -Longer shelf-life : . ; . .
i -Seed variety i -Traceability i delrvery i -Traceabrlrty
L ..................... L ......................... =. .-.T_t?.cr~'abi.l.itr .............................
3 -Quality 1 . . .
: . . : -Skrlled growers : . . : -Safety
Intangible E ugspectron . 5 -Quality certification E -Skrlled handling 5 guaranteed
. - - afety Inspection . . . - -Trmely delrvery . . .
Trans 3 -lm roved 5 -Safetycert1ficatron : -Brand su 0 rt : -Mrnrmum
; emfiency : -Reliable quality g pp 3 shelf-life
......................................................................... HS'S'CEHHHWH'
i Traditional
....................... Market
. -Cost of inputs ;Ma;:;t::;e -Road picked -Low cost
Tangible : -Cost of services : pp : -Poorly packaged : . .
Traits 5 -Traditional ‘ 'L0we'°°" i -Shorter shelf-life * 'M'xed qua"? i
; varieties E -Shorter shelf-life E -No traceabili E -No traceability E
‘ ...................... L .2139. 59999120!!! ________ i _______________ ii ..... i ................... '
. f ; -Unsafe practices 5 -Un rotected
Intangible I i -Unknown quality of l -Sun-exposure i p
. : : : : from hand
Traits . - water - -Long .
: : : . . : touch
. . : transportatrontrme ;

77

Figure 3. 2: Small Farmer Organizations Roadmap to HSSCs

 

Action

 

 

Result

 

 

Agglomeration
compete

Harder market entry
conditions

 

 

 

Forming of
cooperatives

 

 

 

 

 

Emulating larger
farmers

Groups lack
management skills

'— Challenge

 

Responding to specific
market rquirements

 

 

 

 

Networking up and
down the value chain

Increasing efficiency to
expanded networks

 

 

 

Entrepreneurial
decision making

 

 

\V

 

Need to enforce
adoption of stricter
standards

Controlling origin and
use of key inputs

 

 

 

Buyer requirements
met

 

 

 

\l/

 

Thinning out of non

Some small farmers fail ,
compliant farmers

I I I I I I

 

 

I I T I I I

 

78

Sustained market
penetration

*_ 7‘1— T _l_ _l_ *—

 

References

Barham, J. Chitemi, C. 2005. Collective action initiatives to improve marketing
performance. CAPRi working paper No. 74. Collective Action and Property Rights,
October 2008.

Berdegue, J .A. (2001) Cooperating to Compete: Associative Peasant Business Firms in
Chile. PhD thesis. Wageningen University, the Netherlands

Berdegue, J .A., F. Balsevich, L. Flores, T. Reardon. 2005. “Central American
supermarkets’ private standards of quality and safety in procurement of fresh fruits and
vegetables,” Food Policy, Vol 30 Issue 3, June 254-269.

Berdegue, J .A., T. Reardon, F. Balsevich, L. Flores, and R. Hernandez. 2006.
“Supermarkets and Small Horticultural Product Farmers in Central America,” Chapter in
Swinnen, J. (ed.). Global supply chains, standards, and poor farmers, CABI Press.

Calvin, L., Foster, W., Solorzano, L., Mooney, J .D., Flores, L., & Barrios, V. (2002).
Response to a food safety problem in produce: A case study of a cyclosporiasis outbreak.
In B. Krissoff, M. Bohman, & J. Caswell (Eds), Global food trade and consumer demand
for quality (pp. 101-128). New York: Kluwer Academic Press.

Carletto, C. A. de Janvry and E. Sadoulet. 1999. “Sustainability in the Diffusion of

Innovations: Smallholder Nontraditional Agro-Exports in Guatemala. Economic
Development and Cultural Change. 47 (2): 345-369.

Cook, M. and Plunkett, B. (2006) Collective entrepreneurship: An emerging phenomenon
in producer-owned organizations. Journal of Agricultural and Applied Economics 38 (2):
421-428.

Comforth C. (2004). The governance of cooperatives and mutual associations: A paradox
perspective. Annals of Public and Cooperative Economics, 11-32.

Downing, J ., Field M, Kula, O. 2006. Globalization and the small firm: a value chain
approach to economic growth and poverty reduction. USAID Accelerated
Microenterprise Advancement Project. C02

FAO (2006) Promoting Sustainable Producer Group Enterprises: A review of FAO
experience (1981-2006).

Farina, E. and T. Reardon. 2000. “Agrifood grades and standards in the extended
MERCOSUR: conditioners and effects in the agrifood system,” American Journal of

Agricultural Economics, 82(5), December: 1170-76.

Faulkner, D O. and De Rond M. (2000). Cooperative strategy: Economics, business, and
organizational issues. New York: Oxford University Press Inc.

79

 

u?

Flores, L. 2002. Labradores Mayas: A successful experience in the supply of fresh
vegetables to supermarket chains. RIMISP award winning paper. Chile, RIMISP.

Gray, T. and Kraenzle, C. (2002). Problems and issues facing farmers' cooperatives. RBS
Research Report 192, Washington, DC: USDA.

Guba, Egon G. and Lincoln, Yvonne S. 1989. "Judging the quality of fourth generation
evaluation." In E.G. Guba and Y. Lincoln. Fourth generation evaluation. Newbury Park,
CA: Sage, pp. 228-51.

Guba, E. G., & Lincoln, Y. S. (1994). Competing Paradigms in Qualitative Research (pp.
105—117). Handbook of Qualitative Research, edited by N. K. Denzin & Y. S. Lincoln.
Thousand Oaks: Sage Publications.

Hart, 0., and Moore, J. (1990). Property rights and the nature of the firm. Journal of
Political Economy 98(6):] 119-1 158

Helling, J ., Lundy, M., and Meijer, M. 2006. Farmer organization, collective action and
market access in Meso-America. CAPRi working paper No. 69. Collective Action and

Property Rights.

Hernandez, R., T. Reardon, and J .A. Berdegue. 2007. "Supermarkets, Wholesalers, and
Tomato Growers in Guatemala," Agricultural Economics, 36(3), May.

Jano, P., Mainville, D. 2006. Public & Private Roles in Promoting Small Farmers’ Access
to Non-traditional Markets: Case Studies from Central America. IAMA submission 2006
Conference. Buenos Aires, Argentina

James J .W., Glenn H. Sullivan, and Guillermo E. Sénchez 2000 “Future market
development issues impacting Central America’s nontraditional agricultural export

sector: Guatemala case study.” American Journal of Agricultural Economics. 82: 1177——
1183.

Jano, P., Lundy,M., Balsevich, F., Berdegue, J ., Reardon, T. (2004). “Farrners’
organizations and supermarkets in Central America”, Report. Rimisp — Latin American
Center for Rural Development, Santiago, Chile. Jones, E. 2004. Wealth-based trust and
the development of collective action. World Development 32 (4): 691—711.

Maxwell, Joseph. 1996. Qualitative Research Design. An Alternative Approach.
Thousand Oaks: Sage Publications, pp 36-45.

Maertens, Miet and Johan Swinnen. 2007. Trade, Standards and Poverty. Licos Centre
for Institutions and Economic Performance and Department of Economics, University of

Leuven, Belgium.

Mainville, D. Y. and H. C. Peterson (2005). "Fresh Produce Procurement Strategies in a

80

 

Constrained Supply Environment: Case Study of Companhia Brasileira de
Distribuicao." Review of Agricultural Economics 27(1): 130-3 8.

Mainville, D. Y. and D. Zylbersztajn (2005). "Determinants of Retailers' Decisions to
Use Public or Private Grades and Standards: Evidence from the Fresh Produce
Market of Sao Paulo Brazil." Food Policy 30(3): 334-53.

Marin, TR. 2008. Personal, in-depth interviews at CENMA, Guatemala City.

Muradian, R. Mangnus, E. 1999. The Challenge of Entrepreneurship in Agricultural
Cooperatives. http://www.thebrokeronline.eu/en/Dossiers/Special-report-The-power-of-
value-chains/The-challenge-of-entrepreneurship-in-agricultural-cooperatives.

Okello, J .J ., and SM. Swinton 2007. “Compliance with International Food Safety
Standards in Kenya’s Green Bean Industry: Comparison of a Small and a Large Scale
Farm Producing for Export.” Review of Agricultural Economics 29(2): 269-285.

Ortmann, G., and King, RP. 2006. Small-scale farmers in South Africa: Can agricultural
cooperatives facilitate access to input and product markets? Department of Applied
Economics, College of Agricultural, Food and Environmental Sciences. Staff Paper P06-
4. University of Minnesota.

Pomareda, C. 2001. Small Farmers and their Participation in Central-American
Agricultural Exports. UNCTAD Regional Workshop Proceedings on the Regional
Integration and International Linking for the Agro Food Sector Development. San Isidro
de Coronado, Costa Rica.

Porter, ME. (1991) "Towards a Dynamic Theory of Strategy", Strategic Management
Journal, 12 (Winter Special Issue), pp. 95-117.

Reardon, T. and E. M. M. Q. Farina (2001). The Rise of Private Food Quality and Safety
Standards: Illustrations from Brazil. International Food and Agribusiness

Reardon, T. and J .A. Berdegue. 2002. The Rapid Rise of Supermarkets in Latin America:
Challenges and Opportunities for Development. Development Policy Review 20 (4),
September, 317-34.

Rubin H. and Rubin I. 2005. Qualitative Interviewing. The Art of Hearing Data.
Thousand Oaks: Sage Publications.

von Braun, J. Hotchkiss, D., and Immink, M. 1989. “Nontraditional Export Crops in
Guatemala: Effects on Production, Income, and Nutrition,” Research Report no. 73
(International Food Policy Research Institute, Washington, DC, 1989).

Yin, R.K. (1989). Case Study: Design and Methods. SAGE Publication, Thousands Oaks,
California.

81

 

CHAPTER 4
SMALL FARMER’S HOUSEHOLD PHYSICAL, HUMAN, NETWORK AND

ENVIRONMENTAL ASSETS AS EXPLANATORY VARIABLES IN FARMERS
ACCESS TO HSSC’S

4.1 Introduction

It has been widely recognized that small farm households are not necessarily
homogeneous units of decision making. Profound changes in the opportunities for
earning income on the farm, either through improved technology or through different
choice of markets, may have implications for a number of household activities such as
division of labor and budget allocations (von Braun et al., 1989). Assuming that all
farmers seek to increase their incomes through better paying markets, this study assesses
the probability of farmers’ success in penetrating high-standards supply chains (HSSCs)
as a function of the aggregate effect of their asset endowment. The word asset is used
here to describe farmer physical, human, network and environmental characteristics
related to their production and marketing activities.

This essay focuses on the Guatemala lettuce sector drawing mainly from the
literature on the role of small farmers’ assets in adopting non-traditional export crOps
(NTAX) in Guatemala (von Braun et al., 1989, Barham 1995 and Carletto et al.,
1999).The research also draws on the findings by Balsevich (2006) and Hernandez et al.,
(2007), Okello (2006) and Barham and Chitemi (2008) on determinants of supermarket
channel choice by small farmers carried out in Nicaragua and Guatemala, Kenya and
Tanzania respectively. These authors provide insights on farmers producing fresh fruits
and vegetables where the role of small farmers’ assets in improving their market

performance has been assessed using production functions. Several similarities faced by

82

 

lettuce producers to supply HSSC’s and the groups studied by this previous research can i;-
be outlined. Particularly, it is important to acknowledge that larger farmers are better
suited to compete than asset-poor, small farmer groups in face of increasing standards
required by better-paying HSSCs.

This paper contributes to the development debate on the role of farmers’ assets in
their ability to access markets by addressing the following research question:

0 What physical, human, network and environmental assets have enabled small-
scale lettuce farmers to enter HSSCs?

The number of variables tested in the sample population provides a simpler and
broader perspective on how to test the effects of farmers’ assets on small farmers’
capacity to enter HSSCs. To this effect, a descriptive analysis provides the major
differences between household demographics, land and other key physical assets between
suppliers to HSSCs and those still participating in traditional market channels. A second
stage of more sophisticated quantitative analysis is a probit regression where variables
shown as strong determinants of market channel choice in previous studies are run as
independent variables. As a result, identification, description and correlation of key
assets with HSSCs market penetration provide complementary results to understand the
strength of association of variables with access to HSSCs individually as well as in their
aggregate effect.

This paper is presented in four additional sections. Section 4.2 provides a brief
description of the Guatemala lettuce sector and why small farmers are good study
subjects to apply quantitative methods to determine the role of farmer household assets in

HSSCs participation. Section 4.3 describes the theoretical framework outlining the

83

 

farmers’ assets and characteristics most likely to provide weak and strong explanatory
power to the capacity of accessing HSSCs. Section 4.4 provides an account of the
methods used to analyze the data and the corresponding data collection techniques and
use of geographic positioning system tools to verify key farm characteristics. Section 4.5
summarizes the results of the research while section 4.6 offers conclusions and

recommendations for future research.

4.2 Prior market booms and the Guatemala lettuce subsector

Worldwide markets for fresh fruits andvegetables have grown exponentially over
the past three decades, both in volume and diversity of demand and in sources of supply.
In the case of Central America—and other emerging economies—this growth has
evolved from non-traditional agro-exports (NTAX) to the increased importance of
domestic and regional HSSCs, particularly supermarkets (Reardon and Berdegue 2002;
Reardon and Flores, 2006; Reardon and Timmer, 2006).

Guatemala’s agriculture sector is generally characterized as a dualistic system
with a dynamic large farm sector and a stagnant but persistent micro farm sector
(Williams 1986). Prior to the late 19705 Guatemala was an important player in land and
capital-intensive products that enjoyed several market booms (mainly in commodities
such as coffee, bananas and cotton). As such, these crops have historically favored
participation of larger farm holdings. By the early 19805, the development of NTAX had
become important in the development literature as this market boom favored the
inclusion of the small farmer households (Williams 1986; Barham et al., 1995; Carletto et
al., 1999; James et al., 2000). Since the 19805 and through the 19905 development

scholars were curious about the effects of NTAX to reduce poverty by increasing small

84

farmer household incomes (Barham et al., 1992; Carter 1996; James et al., 2000). Thirty
years later, authors maintain that NTAX have had—and continue to have to present
date—positive effects on the Guatemalan economy, small farmer inclusion and on
poverty reduction (Lundy 2007; Carletto et al., 1999). However, negative effects have
also been studied. Of particular importance was the risk of small farmers’ jeopardizing
their ability to produce their own food staples (maize and beans) by growing highly risky
export crops instead. Another concern was the increasing soil toxicity levels over time
due to high amounts of pesticide application on export crops (von Braun et al., 1989;;
Thrupp 1995; Hamilton et al., 2001; Hamilton et al., 2003; Unhever 2003; Hamilton and
Fischer 2005).

Similar to the benefits of NTAX on small farmers, the production of lettuce in
Guatemala is concentrated in the highlands where the agro-climatic characteristics allow
year-round production of temperate-weather crops. The highest concentration of
producers for traditional and HSSCs is clustered in the departments of Guatemala,
Sacatepequez and Chimaltenango, with other clusters in the departments of Solola and
Quetzaltenango as shown in Figure 2 (INE 2003). These areas produce most of the
lettuce consumed in Guatemala and El Salvador, with a small fraction shipped to
Honduras a few months a year. El Salvador and Honduras have only a few areas with
ideal agro climatic conditions to grow lettuce year-round (Granados 2004).

The most recent national survey of the agricultural sector reports about 1400
small and medium farmers growing lettuce in Guatemala (National Statistics Institute —
INE - 2003). The survey shows that over 97% of Guatemalan lettuce producers own less

than 2.5 hectares, except for four large farming companies with total land ownership of

85

 

20 Ha. Triangulated information between area planted, trade data for the Central L '
American market and secondary data on the fruit and vegetable sector point out a total
country lettuce output of circa 13783MT (INE 2003; Granados 2004; CENMA 2008,
MAGA 2008). The lettuce production statistics and lettuce value chain map for 2003—
2004 are presented in Table 4.1 and Figure 4.1 respectively.

The lettuce sector was chosen for this study because it has two important
characteristics with respect to the effect of small farmer assets in the development debate.
First, in contrast with commodity-type products such as tomatoes, lettuce is a specialty
with a far more complex supply chain where product requirements in terms of post
harvest handling such as cooling and retail packing are more challenging to resource-
scarce small farmers. A number of tangible and intangible characteristics required to
supply lettuce to HSSCs are presented in Figure 4.2. Second, lettuce suppliers are mainly
small scale producers located in the Guatemalan highlands, an area with some of the
highest poverty and illiteracy levels in Latin America (UNDP 2007). Studying farmers
that have accessed HSSCs consistently over the past ten years offers a fresh perspective
in understanding the evolution of market procurement structures. These requirements
represent surrnountable challenges for some small farmers already accessing HSSCs, but
are insurmountable challenges for others.

In 1989, von Braun et al., published the first study to measure the effect of assets
in adopting NTAX production using data from Guatemala. In 1985 data gathered in
highland villages with a high number of members in the NTAX cooperative Cuatro
Pinos3 were selected. At time gaps of approximately five years in between, these same
villages were subject to other studies evaluating NTAX adoption, land accumulation

3 . . ,, .
Cuatro Prnos translates to “Four Pines. A cooperative

86

patterns and also income effects (Barham et al., 1995; Carletto et al., 1999; Carletto etal.,
2009). The economic analysis implemented and results obtained up to 1999 by these
authors provided the basis to evaluate a more recent market boom—supermarkets—in
Nicaragua and Guatemala with similar production functions (Balsevich 2006; Hernandez
et al., 2007). This was based on the rationale that as supermarkets increased in market
share and adopted business practices of their multinational counterparts, this market
channel became an option for small farmers like in the early days of NTAX development.
Supermarkets then implied a set of challenges to small farmers to access that are
important to the debate on global trade and its effects on poverty reduction. A detailed
account of the contributions to the literature on farmers’ assets and their effect on market
channel choice is detailed ahead as part of the background on how this study expands the
previous analytical approaches on farmers’ assets and access to HSSCs (supermarkets

and foodservice market channels).

4.3. Literature review
4.3.1 The role of assets in NTAX adoption—learning from previous research
Using a probit regression model, von Braun et al., (1989) assessed the
determinants of small farmer NTAX adoption using data collected in 1985. Using a
production function different types of assets were regressed against the independent
variable represented by the binary choice to farmers between growing NTAX or
traditional staples production. The production function was represented by a vector of
quasi-fixed capital assets (land quantity and quality); a proxy for transaction costs
(distance to paved roads); proxies for risk were included as a set of farmer context

specific shifters such as education and age of the head of household and household on/off

87

 

farm employment. Their results showed that the NTAX boom had shifted the structural
pattern of small farmer access to land, increasing the ownership of this asset in
households that adopted NTAX with the least amount of initial land area. These results
offered strong evidence of one type of linkage which can arise between adoption of high
profit-generation crops and land accumulation patterns.

Motivated by the results of von Braun et al., (1989), Barham et al., (1995), carried
out an assessment of land accumulation patterns and adoption of NTAX in the same
villages4 five years after the first assessment. Chosing a tobit model, Barham et al.,
analyzed the phenomenon with a different constellation of variables. Some variables were
specified the same as by von Braun et al., but more specific proxies for input and output
costs and context-specific shifters were added to capture the adoption effects. For
instance, instead of distance to roads distance to product gathering and processing centers
was regressed. Two variables were introduced in the work of Barham et al., as specific
shifters. The religion shifter was added after an article in the University of Texas Press by
Sheldon (1987 quoted in Barham et al 1995) suggested that Protestant evangelicals were
more likely to adopt NTAX than Catholics. This variable was operationalized as
Catholics or Protestants. The second variable was a qualifier for different microclimates
as potential determinant of NTAX adoption.

The results of Barham et al., (1995) validated the important role of the land asset
found by von Braun et al., (1989), but also showed detailed insights on the dynamics of
this farmer characteristic. For instance, farms as small as 1.2 hectares were estimated to

have a 73% probability of growing some NTAX products; nonetheless, these producers

 

’ The major villages were Pachall, San Jose’ Pacul. Santa Maria Cauqué, San Mateo Milpas Altas, El Rej6n,
and Santiago Sacatepequez.

88

appeared to hit a ceiling to their extent of adoption. Additionally, the researchers found 1
that for farms with two to four hectares the expected land dedicated to NTAX production
leveled off at about 0.35 hectares before beginning to rise again for larger farms. This
situation was explained in part through the econometric analysis suggesting that financial
constraints underlie the small farm adoption ceiling, rather than exhaustion of family
labor supplies—an effect also found in this study for lettuce as explained in the results
section ahead and supported by both ANOVA and econometric assessments.

The distinction between access to land and ability to adopt capital-intensive crops
has been critical to understand NTAX adoption. Often, the presumed necessity of family
labor in the production of labor-intensive crops has provided the basis for optimistic
projections regarding the viability of small farmer participation in modern market
channels (Barham et al., 1995). However, households with abundant family labor often
portrayed as having a comparative advantage in agricultural diversification projects may
not lead to access to NTAX if the financial support to acquire inputs and services is not
provided (Barham et al., 1995).

The explanation of this complex correlation between family size and capital
investment with adoption of NTAX has been critical to the survey questions in this study
for two major reasons. First, household size often measured in number of family
members can be misleading and was broken down into total number of family members
and number of non-child family members (age 15 and older) to capture more detailed
dimensions of family labor availability. Second, land as the major physical capital asset
is often considered a one-dimensional variable measured in total land owned. However,

in highly rugged geographic areas such as the Guatemalan Highlands, land tenure

89

 

F..- ‘Y I

dynamics are complex with different land renting dynamics given the production of
several cropping cycles per year for a variety of products such as lettuce with a
phonological cycle of seventy days. Therefore, survey questions related to the land asset
elicited the land use dynamics in as much detail possible. The investment in doing so
provided good prediction results as discussed ahead.

The increasing importance of the NTAX boom in Guatemala motivated several
other studieson different effects of NTAX adoption. One of these additional studies was
carried out by Carletto et al., (1999) who attempted to model adoption or withdrawal
from NTAX production over time. Ten years after von Braun et al., (1989) collected data
from members of the Cuatro Pinos cooperative, Carletto et al., studied this population
with a more complex production function. This included an increasing number of
variables and some of the same variables with modified forms of operationalization. The
econometric analysis was also modified by first estimating a probit model of the binary
choice of whether farmers became members of the cooperative and then estimating a
tobit model to account for the extent of adoption.

Among the variables modified, the land tenure was divided between area devoted
to NTAX and area devoted to traditional crops. Also, both soil moisture-holding capacity
(humidity) and irrigation were used as indicators of soil quality. Perhaps the most
complex of the new variables introduced by Carletto et al., (1999) was the concept of
time. By 1999 the effect of time (years of experience with NTAX) in the same villages
where data was collected in 1985 by von Braun et al., was specified as historical time and
its effects on the learning process. Additionally, membership in the cooperative was also

introduced as a shifter since not all suppliers to Cuatro Pinos were members. A set of

90

additional variables concerning the role of village effects was included in terms of

distance to markets, infrastructure and available services to farmers.

4.3.2 Modeling supermarket channel adoption after the NTAX experience

Following suggestions from qualitative studies pointing to the importance of the
rise of supermarkets in Central America, similar model estimation methods used in
NTAX were developed to estimate market adoption models in NTAX (Balsevich 2006;
Hernandez et al., 2007). Data gathered in 2004 on the tomato sectors of Nicaragua and
Guatemala respectively used a probit model to measure the determinants of farmers’
market channel choice between the supermarkets channel and the traditional (wet) market
channel. The dichotomy in market channel choice involved significant differences in
quality and safety standards, particularly imposed by supermarkets. The general problem
statement in these studies was that complying with supermarket requirements involved
having the minimum physical assets (land and land quality), certain costs of inputs and
outputs (with distance to markets as a proxy) and other key context-specific shifters
(Balsevich 2006; Hernandez et al., 2007). Following the trend from previous NTAX
adoption models, some innovations to the supermarket channel adoption model were also

added in terms of proxies to wealth such as ownership of vehicle and livestock.

4.3.3 Empirical evidence from Kenya guides analysis on HSSCs adoption

Assessing the capacity of Kenyan small farmers to join NTAX, Okello (2006)
approached the market channel adoption model from a slightly different angle than the
previous studies. Instead of depicting the problem from the supply side as a market

channel adoption decision by the farmer, the research question was posed in the opposite

91

 

 

direction. In other words, the research question dealt with what assets the farmer
possessed that the market channel decision maker (in this case an exporter firm) looks for
to increase his market access performance. Being chosen for that market is egates to
landing a contract. The author conducted this analysis by using a probit regression in the
first stage to measure the effect of small farmers’ capital endowments on the probability
of being contracted. A second stage, a Poisson regression, then measured the effects of
capital endowments on the ability of small farmers who obtain a contract to comply with
strict international food safety standards (Okello 2006).

The model specified by Okello (2006) introduced important changes in the
regression of physical assets. He included the testing of three forms of wealth divided as
land wealth, physical wealth other than land, and farm income (farm and off-farm). Other
variables typically regressed in previous empirical research were irrigation (a proxy for
land quality), human capital characteristics such as age, education and experience in
green bean production.

Okello’s results suggest important considerations concerning buyer-supplier
relationships where green bean suppliers expect to be given a contract based on their
ability to fulfill specific standards. This also exist in the lettuce value chain, particularlyin
HSSCs. Until the mid 1990’s, suppliers wereless concerned with applying stringent
standards to domestic markets even when stricter controls on pesticide residue levels
were being imposed by US and European regulatory agencies on export products (Trupp
1995). This changedas consumers advocated for higher quality and safety, leading to
more stringent public and private standards in retail and foodservice sectors (Farina and

Reardon, 2000; Calvin 2002; J affee 2004). Dynamic markets sought this opportunity to

92

diversify from the traditional wet markets giving origin to the rise of HSSCs where
market channel buying agents many times act as standard enforcers in countries where
public standards are almost non-existent (Berdegue et al., 2006).

An argument often encountered during interviews with key informants was that
HSSCs have enjoyed increasing numbers of farmers of all scales interested in supplying
them. This has helped HSSCs adopt stringent quality, safety and logistics standards
which are key to keep transactions costs as low as possible to compete. As a result,
suppliers comply with requirements in order to avoid losing market share to other
suppliers. HSSCs buyers find themselves in a “buyers’ market” with the chance to choose
among many interested suppliers willing to adjust their business practices for the
opportunity to engage in long-term business relationships with a more profitable option

than traditional markets (Marin 2008).

4.4 Theoretical model

Market channel ‘adoption” is modeled drawing on the usual form of the behavioral
function used to model technology adoption. Balsevich (2005) and Hernandez et al.,
(2007) have considered this appropriate as the market channel choice is essentially a
‘post-harvest technology’ decision following studies by Goetz (1992) and Holloway et a1.
(2005). Feder et a1. (1985) model this decision as a function of a set of incentives facing
and capacities of the farmer, similar to the input demand function derived from profit
function without requiring the assumption of profit maximization wherein adoption is a
function of relative prices of outputs and inputs, risk, and a vector of quasi-fixed capital
assets, as well as various context-specific shifters (Sadoulet and de Janvry, 1995). As this

function is modeled, the technology correlates of channel adoption by modeling the

93

production functions of the two groups of producers, to the traditional market, and to

HSSCs, is analyzed (Hernandez et al., 2007).

4.5 Empirical model specification

Drawing on the experience of previous similar studies, the probability of farmers
participating in one market channel versus another can be measured using a probit model
(Barham et al., 1994, Carletto et al., 1999; Balsevich 2006; Okello 2006 and Hernandez
2006). This binary choice model was chosen as it allows interpreting how a unit increase
in the dependent variables affect the probability of a farmer to enter the market channel
choice or independent variable (Barham et al., 1994; Sadoulet et de Janvry 1995; Carter
et al., 1996; Carletto et al., 1999, Balsevich 2006; Okello 2006; Hernandez 2007). The

probit model is represented by
( 1 )

G( z)= (I)(z) 21-00 (p ( v)dv
Where

(p (v) represents the standard normal probability distribution

(2) (277)”2 eXp"’2/2’

The probit model then measures the producer’s probability (0, l) of participation in the
HSSCs under the following characteristics:

(3) Pr(yi=1|X) = ¢(Xb)

Where i represents the producer and X represents the vector characteristics affecting
producers such as input prices (distance to roads as a proxy); farmer specific variables

(represented by human assets), network and institutional variables (represented by

94

network assets), and farmers quasi-fixed capital variables (represented by physical and
environmental variables). (I) is the standard cumulative normal probability distribution
and xb is the probit score or probit index. Building on the findings of previous studies
outlining relevance of different assets, this probit model equation can be also expressed

by category of assets as follows:

(4) Probability of accessing HSSCs = Market channel choice (Pa, Ha, Na, Ea) + 9,

Where Pa represents physical assets, Ha represents human capital assets, Na
represents network capital assets and Ea represents environmental assets. Each of the
variables is explained below. A hypothesized effect on accessing HSSCs is represented in
the parenthesis preceding the variable name and the information in parenthesis at the end
explains how variables where measured. The a priori effect is derived from the reviewed
theoretical background and field information gathered through qualitative interviews in
the recently completed value chain assessment and case study of Labradores Mayas
shown in previous chapters and summary tables. The explanation of explanatory

variables is shown as follows:

Where pa represents physical assets:

(+) Land owned (in hectares)

(+) Total land worked (in hectares)

(+) Average lettuce plot size (in hectares)

(+) Number of cycles produced (number harvest in the same field per year)

95

(+) Owns a vehicle (dummy for ownership of car or truck 1: yes/ 0: no)
(+) Owns livestock (dummy for bovine or equine livestock 1: yes/ 0= no)
(+) Drip or sprinkler irrigation equipment for lettuce (dummy for drop equipment
ownership 1= yes/ 0= no)

Where Ha represents human assets:

(-) Age of head of household (# of years)

(+) Education of head of household (years of school attendance)

(+) Household size (all members living in household)

(+) Household members working on farm (adults 15 years+)

(-) Off-farm family labor (man days/year)

(+) Farm hired labor for lettuce (man days/year)

(+) Experience growing lettuce (number of years)

Where Na represents network assets:
(+) Membership in grower association(s) (dummy for l= yes/ 0= no)
(+) Total number of upstream input and service suppliers (# of suppliers);

(+) Total number of downstream buyer (# of buyers)

Where Ea represents environmental assets:

(+) Availability of water in the farm (has a well yes or no)

(-) Altitude (meters above sea level)

(+) History of hail damage during years 2003-2008 (dummy for 1= yes/0= no)

(+) Distance to paved roads (km)

96

And e is the stochastic (error) term.

4.6 Data and estimation

A standardized questionnaire was developed to collect data from a statistically
representative sample of the lettuce farmers’ population limited to the communities of
Patzicia, Tecpan Guatemala, Santa Maria Balanya, Saragoza, Patzun and Comalapa all in
the department of Chimaltenango (Figure 4.3). These villages represent the major lettuce
catchment area of HSSCs and traditional markets. Other communities in the western
highlands, particularly the valleys of Zunil and Almolonga, produce lettuce for the
traditional markets for the Central American region; however, since HSSCs do not source
lettuce from those valleys, the sampling was limited to the aforementioned villages to
control for comparable distance to paved roads and markets.

The sampling frame entailed a two-stage stratified random sample among those
accessing HSSCs as their major market and those accessing traditional markets. First,
lists of farmers accessing both market channels were developed in the target geographic
areas obtained from farmer organizations and through key interviews with input and
service providers and wholesalers. The lists of growers were crosschecked before the
survey started until interviews showed it was not possible to add more growers to the two
market channel strata. A total of 627 farmers were identified from which 271 participate
in farmer associations and the rest in two other farmer entrepreneurial groups. The rest of
the producers identified (356) supply the traditional markets through roadside
transactions, terminal markets and through informal contracts with wholesalers. A
random sample of each stratum was taken at 95% confidence level and 50% response

distribution to get a maximum sample size. For the supermarket channel 151 completed

97

questioners were obtained, while 176 completed questionnaires were obtained for the
traditional market channel. Potential coverage errors were more likely in the traditional
market stratum where at least 56 subjects not found in the second attempt to interview
them were replaced by redrawing replacement from the pool not previously selected. A
total of 327 questionnaires were completed among the two strata.

The likelihood of sampling observations is dependent on a choice made by the
subject of study which is in itself the dependent variable. This is called endogenous
stratification and it provides the undesirable likelihood of interviewing more subjects of
the dominating strata more than the other, hence the endogeneity. Since endogenous
stratification is likely, a Heckman (1978, 1979) two-stage method was used to address
the selectivity bias and control for the conditional probability of a supplier being in
HSSCs or in the traditional channel. The procedure consists of the estimation of the
production function for each stratum of producers. To control for the conditional
probability of a farmer being in a particular (endogenous) stratum, the inverse Mills ratio
(IMR) is included as a regressor in the production function regression. The IMR is
calculated (for each farmer) from the market channel regression. In similar studies, when
using the Heckman procedure, the variables in the second stage, used here as the
production function, belong to a bigger pool of variables that are used for estimation in
the first stage (the market channel adoption function); these variables are not reported in
the first stage results, since
they are used as controls (Hernandez et al., 2007).

An important innovation in collecting this data was the recording of GPS-

coordinates on all randomly selected subjects interviewed using a Garmin GPS e-Trex

98

model with an error of five to seven +/- meters. Any of the northernmost comers of the
lettuce farming plot where the grower was found was recorded. From the GPS reading
three variables were recorded (altitude, latitude and longitude). Only altitude was
regressed in the econometric analysis as proxy for agro-climatic conditions to grow
lettuce. The inclusion of altitude was based on previous interviews with farmers and
buyers described the area of Chirijuyu and Patzicia as to what were the “ideal” conditions
to grow lettuce. Based on their responses, the rational on using this variable was that
temperature is positively correlated with altitude. Longitude and latitude were not
regressed, but were used to cross-check distances to major geographic points such as
paved roads, nearest village downtown, wholesale markets and health centers.

The probit estimation model was run with twenty-one variables representing the
four categories of assets under study. Physical and network characteristics were selected
based on theoretical background concerning their explanatory power of adoption models.
Previous empirical research for market channel adoption assessments have pointed out
the problem of endogeneity leading to biased estimators (Barham et al., 1995). A
practical way to overcome this problem was to avoid regressing independent variables
highly correlated with the each other and also variables highly correlated with the error
term. Testing specific variables for endogeneity using the Hausman endogeneity test was
the alternative chosen for this study. The use of the Hausman test was used in two stages.
In the first stage, an OLS regression is run with the chosen instrumental variables to

generate residuals. In a second stage, the residuals are run as an additional variableS . The

’ Smith and Blundell (1986) proposed a test of exogeneity for a probit (tobit) model (1986). The test
Involves specifying that the exogeneity of one or more explanatory variables is under suspicion. Under the
null hypothesis, the models are appropriately specified with all explanatory variables as exogenous. Under
the alternative hypothesis, the suspected endogenous variables are expressed as linear projections of a set of

99

Instrumental variables are exogenous variables to the adoption model but that also
explain in part the adoption decision. In other words, instrumental variables were not
choices made by the farmer to enter the target market channel. In this case, kilometers
from the farm to the paved road and household size in number of members were used.
This were confirmed exogenous variables to the adoption model and used effectively for
the Hausman test. The Hausman test was performed on all physical and network assets
following what other studies have done (Barham et al., 1995; Okello 2006). The variable
lettuce grown (number of cycles multiplied by average plot size) was dropped as it was
discovered to be endogenous. The results for the physical and network asset variables
used show no evidence of endogeneity. Three examples of Hausman tests tables are

shown in Table 4.2 and 4.3 and 4.4.

4.7 Results

This study found important evidence in support of the role of physical, human,
network and environmental assets in determining small farmers’ access to HSSCs. In
general, the results are supportive of previous studies and for some specific assets they
offer different views on the level of importance for HSSCs Second, regression results
are reported for the four asset categories in the estimated model and the lessons learned

on the empirical model estimation.

4.7.1 Descriptive statistics of physical, human, network and environmental assets

Physical assets

 

instruments, and the residuals from those first-stage regressions are added to the model. Under the null
hypothesis, these residuals should have no explanatory power. This test is related to an auxiliary regression
test for exogeneity in a regression context, which is considered a convenient alternative to the Hausman test
employed here.

100

The study design placed considerable emphasis on gathering lagged data on
physical assets. As a result, it was possible to compare the difference in physical asset
ownership at three points in time from 2001 (matching the beginning of a fast
supermarket growth), 2004 and 2008 using ANOVA. Tables 4.2, 4.3 and 4.4 show the
results for the three time periods respectively, while Table 4.12 in the appendix
summarizes the significance between physical asset ownership among the two strata of
farmers. An important finding in this study is that, except for land value, sharecropping
and lettuce area planted, the differences in physical asset ownership were already
significant in 2001. This suggests that the importance of physical assets alone does not
influence entrance to HSSCs as the previous literature has shown for other products and
market channels. This could be attributed to the capital-intensive nature of lettuce and the
significant weight of the other assets in supporting access to HSSCs.

Area planted in lettucewas operationalized as the number of production cycles
grown in the same land during a year multiplied by the average size of the plots per
farmer. Lagged data from 2001 analyzed through ANOVA Shows that there was no
significant difference between the two groups. In fact, the mean total area planted per
year was 0.31ha for traditional market suppliers while 0.36ha for HSSCs suppliers (Table
4.4.). However, by 2004 the difference appears to be highly significant at p <0.01. Table
4.3 shows that in about three agricultural years the average total area planted by HSSCs
was 0.44Ha, an 82% increase from the 2001 area planted level. Tables 4.4 and 4.3 show
that, while the land ownership asset and the size of lettuce plots remained almost
unchanged, the number of cycles per year 1.54 to 2.1, a 37% increase from 2001 to 2004.

When comparing 2004 to 2008 ANOVA data, it was confirmed that the land ownership

101

and average plot size assets increased slightly, but the number of cycles went up 40%
from 2.1 l to 2.97 cycles per year.

It is also important to note the differences in land renting patterns throughout
time. While neither farmer group acquire more land in eight years, there was a significant
trend toward land rental. During this periodHSSCs went from 0.10 to 0.41 to 0.45
hectares from 2001, to 2004 and 2005 respectively. In the case of traditional market
suppliers the land renting behaviour was also noteworthy with a five-fold increase from
2001 to 2004 0.05 to .26ha. However, there was a decrease to 0.25 by 2008. There is no
clear explanation for why both groups do not show similar proportions in land renting
from 2004 to 2008. However, it can be inferred that HSSCs remained the most dynamic
land users of the two groups during the 2001—2008 period. For further details see tables
4.2, 4.3, 4.4 and 4.12.

While this information does not argue the importance of initial land assets to enter
HSSCs as shown also in previous market channel adoption models, there is an important
point to make in regards to capital investments needed to access and grow market share
in HSSCs. This brings back the point made previously about financial constraints

underlying the small farmer NTAX adoption ceiling (Barham et al., 1995).

Human assets

It was not possible to obtain reliable information on lagged human assets.
However, important insights are drawn from 2008 data as explained by the ANOVA
results. Concerning age, education and experience growing lettuce, these variables were

Significantly different at the p<0.05 level among the two groups, findings that were

102

nearly mirrored by the econometric results discussed ahead concerning influence on the
dependent variable (access to HSSCs). Table 4.5 provides provides additional evidence
on the degree of available working capital in the HSSCs group with nearly double the

amount of on-farm labor hired at p<0.01 significance level.

Network assets

The ANOVA results show in general highly significant differences in 2008
between HSSCs and traditional market suppliers. This shows the major differences in
network linkages up and downstream in the value chain among the two groups. These
differences are evaluated in the econometric results run only through three variables

(association, total suppliers and total buyers) to avoid endogeneity issues.

Environmental assets

Although Carletto et al., (1999) were the first authors to consider environment-
related variables in a market-adoption regression, few studies consider a broader
evaluation of more such variables. Thisstudy bundled a number of variables that are fixed
by the environment and that were highlighted as important determinants to access HSSCs
in previous qualitative research. The ANOVA results pointed out that there are major
differences between HSSCs and traditional market supppliers in access to water at the
farm, incidence of bail damage and distance to roads. From these three variables, it is
important to consider that the question was intended as a measurement of distance to the
water table or natural surface water as an environmental characteristic. However, answers

to the question revealed that those farmers with water in their farms dug artesian wells or

103

 

had access to deep wells from nearby locations. Therefore, this variable can be
considereda function of capital investments. and not necessarily an environmental asset.
Shall this variable be considered in furture research, it should considered part of the

irrigation equipment.

4.7.2 Econometric results and discussion

In estimating the model for lettuce suppliers accessing HSSCs, it was important to
evaluate the pros and cons faced by previous research in their variable specification, the
expected effect of variables chosen on the dependent variable and the considerations on
continuos expansion of new variables. As a result, the econometric results obtained in
this study are supportive of the general a priori expectations as explained ahead. In order
to keep consistency in the fashion the role of assets have been presented and discussed in
this chapter, the results are offered separating the assets in the four established categories,
physical, human, network and environmental. The regression analysis captures the

aggregate amount of all variables considered. The results are explained as follows.

Econometric results on physical assets

Table 4.8 presents the probit regression results of the seven physical assets
selected after endogeneity tests were performed. As hypothesized, it can be observed that
all coefficients have a positive sign, but only two variables have an effect that is
statistically significant at p<0.01. This is an important finding that supports previous
theory concerning land and capital investments. From the physical assets regressed, it

Was Surprising that irrigation did not appear significant in accessing HSSCs, but there is

104

no possible explanation to why the estimated model does not show that effect. The theory
shows that the estimation of adoption models can produce surprisingly contradictory
results to a priori expectations. The results of Carletto et al., (1995) on similar physical
assets using tobit regression, strongly supported by the theory, were no exception.

This should not be a concern in interpreting the other variables, neither it is
necessary to drop variables to make the model more parsimonious. The model does
support that the average lettuce plot size and the number of cycles produced are
extremely significant which can be considered the closest proxies to the working capital
endowment level of the HSSCs suppliers. The dynamics of land and capital are trapped
by these results considering that lettuce supplied under the tangible and intangible HSSCs

requirements is among the most capital-intensive vegetable crops grown.

Econometric results on human assets

As hypothesised, education of head of household is significant (p<0.05) while
experience in growing lettuce is also significant at p<0.10 level. In can be inferred that in
a market such as HSSCs with increasingly monitored standards, those who can read and
write are in better position to understand written material, and also keeping a logbook of
their cultural practices such as pesticide application. During interviews with farmers, the
need to read and write seemed to have surpassed the basic skill level. In fact, as
procurement requirements increased over time (more than NTAX production requires)
the role of key human such as education may need to evolve to professionalization. This
may lead to estimation modifications in the future such as applying weights to education.

Other alternatives may be to ask more specific questions or deliver a test to capture data

105

beyond years of schooling. Not to belabor the point, the current methods of gathering
data on education are poor since it is impossible to cross check if the subject passed the
years through school in good stand, or if he went to school at all.

The experience if growing lettuce was also significant at p<0.10. Neither
Balsevich (2006) nor Hernandez et al., (2007) regressed this variable in their estimated
models (the closest to the lettuce probit model herein discussed) for Nicaragua and
Guatemalan tomatoes respectively. However, Okello (2006) did regress the number of
years the grower had been producing green beans in estimating participation in a
marketing contract. Okello’s results were also significant and with a positive sign, as the
results show in the present model.

The negative sign of age of head of household confirms a priori expectations of
increased risk aversion to dynamic markets as growers grow older. This was tested
previously by Barham et al., (1995) on NTAX adoption in Guatemala, and Balsevich
(2006) and Hernandez et al., (2007) with significant results in Nicaragua and Guatemala
respectively using similar estimation models. However, Carletto et a1 (1999) did not find
this variable significant in NTAX adoption models. Nevertheless, older farmers tended to
withdraw from adoption confirming the hypothesis with the negative sign and high
Significance level. Based on this result, it is suggested to continue regressing this variable
in filture model estimations as the results may be strongly contextual to the product,
market and site under research.

Also, the negative sign of the farm hired labor variable is a concern that deserves
further evaluation. Even though the econometric results are not significant, it is important

to discuss the farm hired labor variable also from the ANOVA perspective between

106

groups. To illustrate, it was established that HSSCs suppliers hire about twice as much

labor as traditional market suppliers (Table 4.5).

Econometric results on network assets

The three regressed variables show strong significance level in accessing HSSCs
as hypothesized. The signs of the coefficients of membership in associations and access
to downstream links point out the importance of market linkages. During interviews
carried out for the value chain assessment, it was established that farmers with a
diversified portfolio of buyers may be able to sell top quality to the most demanding
markets, while also profiting from linkages to wet markets as outlets for second and third
quality product. This characteristic is strongly captured by the estimation model and hints
at further dissection of the farmer’s network. These variables concerning up and
downstream linkages were regressed for the first time in this study. While that constitutes
a good contribution to the literature, their importance could still be underrated. As in the
case of education, it was Ieamed in chapter one that supply chain network links are
poised to have a relevant, perhaps more important, effect on the ability of farmers to
access HSSCs effectively.

It is important to the discussion to underline that of all the variable groups
targeted, network assets were the hardest to select for the estimation model. This was
because previous attempts in the literature where dimensions of the farmers’ network

were regressed (e. g., religion by Barham et al., (1995)) provided mixed results. A general

107

recommendation for this category of assets in regression models is to avoid variables

. . . . . 6
measured wrth subjective questions and or rnstruments.

Econometric results on environmental assets

The available empirical evidence researched does not provide much basis to
compare and contrast environmental assets in econometric analysis for market access.
This is why this study considers a key contribution the running of important
environmental characteristics pointed out by buyers in their process of selecting the best
suppliers from a pool of farmers in key geographic locations. In this regression the four
environmental variables regressed offer statistically significant results and support the a
priori expectations on their negative and positive effects on the dependent variable as
shown in Table 4.8. More importantly, these results validate the relevance of such assets
for lettuce farmers and in further research dealing with open field agriculture.

These results may have significant implications on the way environmental risks
are assessed prior to planting or sourcing highly perishable products such as lettuce from
areas with a history of hail damage. It should however be noticed that the altitude
variable has a very small coefficient (0.002) and its effect on HSSCs access may not be
relevant in practice. Going back to the data gathered on this variable it was established

that the minimum and maximum altitude for farmers supplying HSSCs was 1969-2364

 

 

6 One of the issues in quantitative research techniques found in the literature of market adoption was how
the variable was measured in the random sample. For instance, the use of dummies like yes or no for
"lembership association easily recorded on dimension of network characteristics. However, other
dimensions like relgion were more troublesome (Barham et al., 1995). This, though, should not deter the
exploration of more variables with explanatory power in determinants of market access modeling. Barham
and Chitemi (2008) appear as the most innovative in the recent literature measuring social structure
variables such as altruism, general trust, help trust and money trust among other variables like group
leadership by sex measured through different group games. While Barham and Chitemi only carried out
ANOVA and Pearson R2 analysis, it would be interesting to test if such variables provide robust and
unbiased estimators in a regression type of model.

108

meters above sea level (a range of 395 meters) while for traditional market suppliers it
was 1923-23 87 (a range of 464 meters).

Collecting and cross-checking distance data through GPS-recording tools was an
important, but time-consuming experience that resulted in little impact to the research.
Nonethelesss, innovation on data collection approaches can be considered part of the
modest contributions of this dissertation. Besides, further studies could enhance the
research techniques through simulation software combining temperature, altitude and
propensity to hail damage using the GPS coordinates and historical data from 2003 to

2008 generated in this research for 327 subjects.

4.8 Conclusions

This study found important evidence in support of the role of physical, human,
network and environmental assets in determining small farmers’ access to HSSCs. In
general, the results were supported by theory, although some assets contradict the a priori
expectations. The econometric results on the importance ofthe four asset categories
regressed in the estimated model and the lessons learned on the empirical model
estimation are concluded as discussed below. The variables used in the regression
provided coefficients with the effect on the dependent variable as hipothesized. Four
important conclusions can be underlined in this study. First, the pool of variables with
robust explanatory power was increased by adding total suppliers and total buyers as key
network assets and by the inclusion of important environmental variables such as hail
damage and altitude. As it has been explained, these variables play an important role in

lettuce production. Further importance of these variables lies in that, even every product

109

will have its own climate-specific requirements that, as in the case of lettuce, will affect
the farmers’capacity to enter HSSCs.

Second, the inclusion of the classic variables that have given strong explanatory
power to the dependent variable in previous studies has been important to this empirical
research. However, important lessons were not realized until the final stages of the data
analysis on the weak manner in which strong variables such as education were measured
by prior studies and this study as well. To illustrate, just asking a subject how many years
he went to school may be necessary, but not sufficient on his level of education. A quick
test may be necessary in order to rank education in more specific dimensions. With the
increasing standards in HSSCs, it can be expected that in future years the capacity of
farmers to use computer-based monitoring programs and farmer tools may be necessary,
and not simply basic reading and writing skills. The same issue in measuring key network
assets applies since belonging to an organization has a considerable explanatory power in
this study and others in the past. However, there are no cross-check mechanisms in the
literature to enrich what this variable means in greater detail. Time and cost pressures
involved in gathering data often leave aside the important role of further verification, but
it is recommended that further research pay attention to these categories of variables.

Third, interpreting econometric results can be limiting as they represent
likelihoods for one particular set of data. This research found useful to supplement the
information on why some variables were significant or not with the means ANOVA. As
such, important findings became more evident among the two groups (HSSCs and

traditional market suppliers) that the regression results do not tell. Good data

110

interpretation practices before regression also allowed the identification of variables with
high probability to be endogenous.

Fourth, this research contributed to the empirical research by presenting a
compendium of variables, some tried several times in market channel adoption models
and some tried for the first time, that were most likely to explain the likelihood of a small
farmer to access a market such as HSSCs. The next step to make these results applicable
in the field would be to standardize these regressions into software tools similar to those
used by banks to calculate credit ratings and capacity to pay of individuals requesting
loans. Developing a tool like that is outside the scope of this study, but would be useful in
guiding practitioners to identify farmers’levels of readiness in acessing HSSCs while
pointing to development efforts specific weaknesses where to allocate support resources
to accelerate market access.

Finally, in interpreting the probit results, greater emphasis was placed on the sign
of the coefficient than on their marginal effect as this results are not definitive, but rather

likelihood obtained for the lettuce sector.

11]

Table 4.1: Guatemalan Lettuce* Production and Value 1989-2006 (000's MT and 000's 5)

 

 

Year Production (MT) Area (Ha) Value (000‘s US$) Average $/Kg
1989 6965 317 291 0.56
1990 6945 316 300 0.61
1991 7976 363 283 0.61
1992 7856 357 314 0.60
1993 8054 366 333 0.61
1994 9056 412 399 0.73
1995 10266 467 429 0.75
1996 10981 499 489 0.65
1997 102 84 467 496 0.74
1998 9297 423 435 0.97
1999 l 1056 503 633 1.02
2000 12463 567 759 1.03
2001 12276 558 764 1.05
2002 13098 595 839 1.09
2003 13783 627 1034 1.09
2004 14830 674 1072 1.10
2005 14010 637 1134 1.14
2006 14715 669 1104 1.28

 

‘ Iceberg lettuce is widely planted in Guatemala. Romaine varieties are also planted upon buyer's request.
Source: Multiple sources (MAGA Central America export records. SIECA, exporting company records).

112

Table 4.2 Mean comparison of household physical asset characteristics in 2008

 

 

 

2008
HSSCs Traditional F - Sig.
Market Statistic

Number of households (a) 151 176
Land-related assets (Ha)

Land ownership 1.19 0.73 17.26 0,000***

Land rented 0.45 0.25 3.31 o,ooo***

Land share-crepped 0.22 0.15 0.41 0053*

Total land farmed 1.86 1.13 43.74 0000*”
Land Value (US$'000/Ha) 33.64 31.12 2.24 0.136
Lettuce area planted 0.59 0.35 21.64 0000*”
Average lettuce plot 0.11 0.09 59.98 0000*"
Number of cycles per year 2.97 1.53 39.18 0,000***
Cycles*average plot size 1.80 0.64 46.87 0000*"

 

(a) Total number of households 327
Mean comparisons between household groups calculated using F-test. *: p < 0.1;
*"‘: p < 0.05; ***: p.< .01

Table 4.3 Mean comparison of household physical asset characteristics in 2004

 

 

 

2004
HSSCs Traditional F- Sig.
Market Statistic

Number of households (a) 151 176
Land-related assets (Ha)

Land ownership 1.1 1 0.71 13.14 0000*"

Land rented 0.41 0.26 1.83 0000*”

Land share-cropped 0.03 0.02 0.02 0.003M

Total land farmed 1.55 0.99 26.05 o_ooo***
Land Value (US$‘000/Ha) 27.12 26.33 0.33 0.569
Lettuce area planted 0.44 0.28 17.02 0,000***
Average lettuce plot 0.11 0.08 54.31 0.000***
Number of cycles per year 2,11 1.60 21.88 0.0004“:
Cycles*average plot size 097 0,50 22.40 0000*"

 

(a) Total number of households 327
Mean comparisons between household groups calculated using F-test. *: p < O. 1;
**: p < 0.05; ***: p < .01

113

Table 4.4 Mean comparison of household physical asset characteristics 2001

 

 

 

2001
HSSCs Traditional F -
Market Statistic Sig.

Number of households (a) 151 176
Land—related assets (Ha)

Land ownership 1.10 0.72 12.08 0000*“

Land rented 0.10 0.05 0.17 0085*

Land share-cropped 0.03 0.01 0.02 0.102

Total land farmed 1.23 0.78 33.32 0,000***
Land Value (US$'000/Ha) 22.24 21.90 0.08 0.780
Lettuce area planted 0.36 0.31 1.84 0.176
Average lettuce plot 0.10 0.09 4.32 0.039M
Number of cycles per year 1.54 1.28 5.25 0,023"
Cycles*average plot size 0.70 0.48 7.08 0.008***

 

(a) Total number of households 327

Mean comparisons between household groups calculated using F-test. *: p < 0.1;

**: p < 0.05; ***: p < .01

114

Table 4.5 Mean comparison of household human assets within groups for 2008

 

 

 

2008
HSSCs Traditional F - Sig.
Market Statistic
Number of households (a) 151 176
Head of household assets (years)
Age head of household 39.33 42.60 9.13 0.003”
Education 5,46 3.96 32.65 .000**"‘
Experience growing lettuce 8,36 6,82 13.25 .000**"‘
Household labor
Household members (all members) 5.50 5.56 0.08 0.772
Household workers 15 years + 3.49 3,46 0.03 0.861
Household off-farrn labor
(days/year) 1905 1956 0.01 0.924

Hired on-farm labor (days/year) 109.38 5932 13,0] 000*”

(a) Total number of households 327

Mean comparisons between household groups calculated using F-test. *: p < 0.1;
“‘2 p < 0.05; *"‘*: p < .01

115

Table 4.6 Mean comparison of household network asset between groups for 2008

 

 

 

HSSCs Traditional F - Sig.
Market Statistic
Number of households (a) 151 176
Membership in associations 0,42 0,19 22.46 0000* * *
Upstream linkages (number)
Input suppliers 1.89 1.28 46.85 0.000***
Transportation service suppliers 0,60 0.15 57.97 .000* **
Credit suppliers 0.60 0.57 0.105 0.747
Packaging suppliers 0.36 0.15 12.51 0000* * *
Other suppliers 0.74 0.20 32.96 0000* "‘ *
Total upstream linkages 4.21 2.36 57.97 0000* * *
Downstream linkages (number)
Supermarket buyers 0.48 0.00 59.24 0000* :1: .1.
Restaurant buyers 0.46 0.00 47.19 0.000
Wholesalers 0.28 0.38 2.99 0850*
Traditional market buyers 0, 53 0,99 8,21 000* * *
Total downstream linkages 1,75 1.36 26,8 5 0000* * *

 

(a) Total number of households 327

Mean comparisons between household groups calculated using F-test. *: p < 0.1;

**: p < 0.05; “*2 p < .01

 

116

Table 4.7 Mean comparison of household environmental asset between groups for 2008

 

 

 

HSSCs Traditional F - Sig.
Market Statistic
Number of households (a) 151 176
Altitude 0.71 0.28 21.07 0080*
Availability of water in the farm 0.68 0.20 18.17 0000*”
Incidence of bail damage on lettuce
Hail damage in 2008 0.22 0.35 7.16 0008*“
Hail damage in 2007 0.38 0.43 0.76 0.383
Hail damage in 2006 0.35 0.42 1.65 0.200
Hail damage in 2005 0.26 0.37 4.66 0.032
Hail damage in 2004 0.23 0.31 2.66 0.104
Hail damage in 2003 0.01 0.03 1.48 0.255
Aggregate (2008-2003) 0.68 0.68 15.61 0000*“
Farm location characteristics (Km)
Distance to paved roads 1.26 1.60 9.93 0002*“
Distance to point of wholesale 9.30 7.35 0.61 0.435
Distance to downtown (input stores) 2.55 2.52 0.02 0.896
Distance to household 1.16 1.02 1.36 0.244
Distance to health centers 2.17 2.29 0.32 0.573
(a) Total number of households 327
Mean comparisons between household groups calculted using F-test. *: p < 0.1; “‘2

p < 0.05; ***: p < .01

117

Table 4.8 Probit results effects of small farmer assets in accessing HSSCs

 

 

markets
p- Marginal
Explanatory variable Coefficient SE value effect Sig.
Constant -1.3183 0.382 .001 ***
Physical assets (pa)
Land owned 0.027 0.033 0.395 0-035
Total land worked 0.278 0.196 0.157 0342
Average lettuce plot size 0.201 0.501 0.000 0202 ***
Number of cycles produced 0.816 0.146 0.000 0.901 ***
Owns a vehicle 0.256 0.329 0.436 0271
Owns livestock 0,220 0,433 0,450 0.321
Irrigation equipment for lettuce 0.208 0.270 0.443 .227
Human assets
Age of head of household -0.004 0.014 0.759 '0-007
Education of head of household 0.157 0.061 0.009 0133 ***
Household size 0.011 0.083 0.899 0-016
Household members working on farm 0.076 0.100 0.444 0-081
Off-farm family labor 0.001 0.005 0.793 0001
Farm hired labor for lettuce -0.001 0.001 0.361 '0-001
Experience lettuce 0.062 0.035 0.075 0.064 *
Network assets
Membership in association 0.646 0.284 0.023 0751 **
Number of upstream linkages 0.023 0.064 0.001 0.043 ***
Number of downstream linkages 0.587 0.196 0.003 0-601 ***
Environmental assets
Availability of water in the farm 1.228 0.002 0.000 1239 ***
Altitude ofthe farm 0.002 0.002 0.088 0902 *
History hail damage 2003-2008 -0501 0.285 0.078 -0-589 *
Distance to paved roads -0.292 0.134 0.029 -0.310 **

 

Significance levels *: p < 0,]; **: p < 0.05; ***: p < .01

118

Figure 4.1: The 2007-2008 Lettuce Value Chain Map

 

Market
Channel
Share

Delivery
Point

 

 

 

Prod uction/
Distribution
Market Share

 

 

 

Sorting
Packing

Wholesale
(Market share)

Production
(% farms)

Production
Origin

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Supermarkets Foodservice Traditional Traditional +
GT/ES GT/ES GT ES
36% 6% 35% 23%
T m a] §"l§zl'arl<'e't's"i i'férini'tfal'i
e n : Nearby 5 : Markets 5
Distribution Processing Markets EProductioni E El :
Center Centers/Rest GTC'ty : Areas E : Salvadori
. I.--- ..... a l-.- .....
. f .3! .
.\.\.\ '\. ’./.l /. !
.. .. _ _. .. .. A- 'x, .. .3. 4.1.1., .. .z' .1...
Matt/41304728} ' l%_]'3% 2%l zo/stvimrso/o} wig/{15%}
l— -" —‘:J\. l -‘. ‘L' . L ‘s l' T4
‘\_ .\ ' ' . ' . ' [I
‘ t75»’-'~’.-.“~. .I'
__-_ :;__ _::b¢__-_

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  
  

' Spot Markets i

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Central Highlands
64% Total Volume

 

 

HSSCs

 

. I I
r--—.s-—-— | I
Wholesaler : Packing I l_92§e"1§l%_.._." 'lr_ELS_%Ed.°£_I
Pack'gFP'am :___P_'afm___' i_§0adsid——: 3133;888:— --
| . I .
Packln shed Lsortln on] 1 '
l. ______ J I. -__§._.y l :
Large Dedicated ’4‘: :
Dedicated Small Farmer Guatemalan Salvadoran :
Wholesalers Organizations Wholesalers Wholesalers “'1'
(12)+12 = 24% 16% 40% 20% :
‘ ..... ‘ ‘ I ‘ :
_.‘.__.:: ............................ {- .... :
Medium farms/ ..... -._.‘._'_c Small farms 5
Wholesalers 88% "T
(12)% 4 :

Western Highlands
36% Total Volume

 

 

 

TRADITIONAL MKTS.

Figure 4.2: Lettuce Tangible and Intangible Product Traits

119

 

l 00%

l 00%

100%

l 00%

100%

Input Production Wholesaling Retailers

Sorting Foodservice
Packing Traditional
Market ......
-Cost of inputs -lmproved varieties fa??? Packa ed
T . : -Cost of services : -Competitive price : ' a e e : ' g
anglble : : . ; -Pre-cooked : -Labeled
T . t : -Approved : -Better handling : -Sam -da : -Cooked
rm 8 5 pesticides 5 -Longer shelf-life 5 d l' e y 5 -Traceabili
: -Seed variety : -Traceability : e Ivery . _ - t)’
' ...................... :. ......................... -_ .'_T!?.C?§l?l'.'!¥ ........ .: ....................
-Quality 5 . . ;
; . . . -Skll|ed growers : . . . -Safety
lntan ible : inspection 5 _ . . . 5 -Skllled handling 5
g . Quallty certlficatlon . -Timely delivery . guaranteed

. -Safety inspection -

 

Traits - I -Safety certification 1 1 -Minimum
: -lmproved ; . . ; -Brand support : .
- efficiency 5 -Rellable quality 5 5 shelf-life
. -Cost of inputs ;Marketable -Road picked -Low cost
Tangible 3 -Cost of services i ppearance 1 -Poorly packaged 1 . .
. . . . : -Lower cost : . : -Mlxed quallty
Traits ;-Tradltlonal ; . ; —Shorter shelf-life ; N b'l' ;
: varieties : -Shorter shelf-llfe : -No traceability : - otracea llty .
Notraceablhty ......................
: : -Unsafe practices .
Intangible 5 5 -Unknown quality of E -Sun-exposure E Unprotected
. - . . . from hand
Traits 5 5 water : -Long : touch

5 transportation time E

------------------------------------------------------------------------------------------

120

Figure 4.3: Research Site

.0"
\
s

‘ Comalapa
‘ o I
I

\ I
Patzun : Santafiruz Balanya:
~ . — a.
l

Tecpan s
. \

I
' ’-

I ‘
~ ~- \ ’ .

Patzicia“? -
l
I
l Zaragoza

~\

 

 

MEXICO

 

 

BELIZE

 

 

GUATEMALA

*Guatemala HOND
City URAS

EL SALVADOR

 

 

Lettuce producing area

Source: Adapted from Guatemala National Institute of Geography (IGN 2004)

121

 

References

Barham, B., Clark, M., Katz, E., Schurrnan, R. (1992). Nontraditional agricultural exports
in Latin America. Latin American Research Review, 27.2, pp. 43-82.

Barham, B., L., Michael R. Carter and Wayne Sigelko. 1994. Agro-export production and
peasant land access: Examining the dynamic between adoption and accumulation. Journal
of Development Economics 46: 85-107.

Barham, J ., Chitemi, C. 2005. Collective action initiatives to improve marketing
performance. CAPRi working paper No. 74. Collective Action and Property Rights,
October 2008.

Balsevich, F. 2006. Essays on producer’s participation in, access to, and response to the
changing nature of dynamic domestic markets in Nicaragua and Costa Rica. Doctoral
dissertation. Department of Agricultural Economics, Michigan State University.

Berdegue, J .A., T. Reardon, F. Balsevich, L. Flores, and R. Hernandez. 2006.
“Supermarkets and Small Horticultural Product Farmers in Central America,” Chapter in
Swinnen, J. (ed.). Global supply chains, standards, and poor farmers, CABI Press.

Calvin, L., Foster, W., Solorzano, L, Mooney, J .D., Flores, L., & Barrios, V. (2002).
Response to a food safety problem in produce: A case study of a cyclosporiasis outbreak.
In B. Krissoff, M. Bohman, & J. Caswell (Eds), Global food trade and consumer demand
for quality (pp. 101-128). New York: Kluwer Academic Press

Carletto, C. A. de Janvry and E. Sadoulet. 1999. “Sustainability in the Diffusion of
Innovations: Smallholder Nontraditional Agro-Exports in Guatemala. Economic
Development and Cultural Change. 47 (2): 345-369.

Carletto, C., Kilic, T., and Kirk, A. 2009. Non-Traditional Export Crops in Guatemala:
Short-Term Tool or Long-Tenn Strategy for Poverty Alleviation? Ill EAAE-IAAE
Seminar ‘Small Farms: decline or persistence.’ University of Kent, Canterbury, UK

Carter, Michael R. and Bradford L. Barham. 1996. Level Playing Fields and Laissez
F aire: Postliberal Development Strategy in Inegalitarian Agrarian Economies. World
Development 24 (7): 1133-1 149.

Granados, J .C., 2004. “Analis del Comercio Regional y Extragional de Frutas y
Hortalizas en Centro America”. PF ID-F&V, Michigan State University. Consultancy
report not published.

Hamilton, Sarah and Edward F. Fischer. 2003 “Nontraditional agricultural exports in
highland Guatemala: understandings of risk and perceptions of change.” Latin American
Research Review 38 (3): 82—110.

122

Hamilton, S. G, Fischer, E. F. 2005. Maya Farmers and Export Agriculture in Highland
Guatemala: Implications for Development and Labor. Latin American Perspectlves,
2005; 32; 33 Latin

Hamilton, S. G. Sullivan and L. Asturias. 2001. Economic and Social Impacts of Non-
Traditional Agricultural Exports in Highland Guatemala: Understandings of Risk and
Perceptions of Change. Latin American Research Review, Vol. 38, No. 3 (2003), pp. 82-
l 10

Heckman, J ., 1978. Dummy endogenous variables in a simultaneous equation system.
Econometrica 46(4), 931-959.

Hernandez, R., T. Reardon, and J.A. Berdegué. 2007. "Supermarkets, Wholesalers, and
Tomato Growers in Guatemala," Agricultural Economics, 36(3), May-

Instituto Nacional de Estadistica. 2003. Guatemala National Institute of Statics
Agricultural Census.

James J.W., Glenn H. Sullivan, and Guillermo E. Sénchez 2000 “Future market
development issues impacting Central America’s nontraditional agricultural export
sector: Guatemala case study.” American Journal of Agricultural Economics. 82: 1177—
l 183.

Jaffee, S. 2004. From challenge to opportunity: The transformation of the Kenyan fresh
vegetable trade in the context of emerging food safety and other standards. Agriculture
and Rural Development Discussion Paper # 1, World Bank, Washington DC.

Lundy 2007. Assessing Smallholder Participation in the French Bean supply chain in
Guatemala Juan Francisco Project. CIAT. October 2007

MAGA 2008. Ventanilla Unica para las Exportaciones. Non-published trade data.
Guatemala.

Okello, J .J . 2006. “Compliance with International Food Safety Standards: The Case of
Green Bean Production in Kenya Family Farms. Doctoral dissertation. Department of
Agricultural Economics, Michigan State University.

Reardon, T. and CR Timmer. 2007. “Transformation of Markets for Agricultural Output
in Developing Countries Since 1950: How Has Thinking Changed?” chapter 55 in R.E.
Evenson, and P. Pingali. Handbook of Agricultural Economics, 3: Agricultural
Development: Farmers, Farm Production and Farm Markets. Amsterdam: Elsevier Press:
2808-2855.

Reardon, T. and L. Flores. 2006b. “ Viewpoint: ‘Customized Competitiveness’ Strategies

for Horticultural Exporters: Central America Focus with Lessons from and for other
Regions,” Food Policy, 31(6).

123

Reardon, T. and J .A. Berdegue’. 2002. The Rapid Rise of Supermarkets in Latin America:
Challenges and Opportunities for Development. Development Policy Review 20 (4),
September, 317-34.

Sadoulet, E., de Janvry, A., 1995. Quantitative Development Policy Analysis. Baltimore:
The Johns Hopkins University Press.

Smith, R.J. and Blundell, RW. 1986. An Exogeneity Test for a Simultaneous Equation
Tobit Model with an Application to Labor Supply. Econometrica, Econometric Socrety,
vol. 54(3), pages 679-85, May

Thrupp, L. A., G. Bergeron, and W. F. Waters. Bittersweet Harvest for Global
Supermarkets: Challenges in Latin America's Export Boom: Natural Resources Institute,
Washington DC, 1995.

United Nations Population Division (UNPD) (2007). Rural population, development and
the environment. Retrieved on 05/01/2009 in
http://www.un.org/esa/population/publications/2007_PopDevt/Rural_2007.pdf.

Unhever, L.J., 2000. Food Safety Issues and Fresh Food Product Exports from LDCs.
Agricultural Economics 23(3), pp. 231-240.

von Braun, J. Hotchkiss, D., and Immink, M. 1989. “Nontraditional Export Crops in
Guatemala: Effects on Production, Income, and Nutrition,” Research Report no. 73
(lntemational Food Policy Research Institute, Washington, DC, 1989).

Williams, Robert G. 1986. Export Agriculture and the Crisis in Central America. Chapel
Hill, NC., USA. University of North Carolina Press.

124

Appendix Chapter 4

 

Table 4.9 Hausman endrgeneity test on total lettuce grown

 

 

Explanatory variable Coefficient p-value
Constant 0.431 0.090
Physical assets (pa)
Total land worked 0.234 0.035
Land owned 0.808 0.160
Average lettuce plot size 0.882 0.000
Number of cycles produced 0.337 0.000
Total lettuce grown“ 0.025 0.037
Residuals* 0.011 0.007
Owns a vehicle 0.351 0.043
Owns livestock 0.936 0.423
Irrigation equipment for lettuce 0.209 0.475
Human assets
Age of head of household 0.289 0.367
Education of head of household 0.872 0.245
Household size 0.351 0.512
Household members working on farm 0.936 0.523
Off-farm family labor 0.234 0.179
Farm hired labor for lettuce 0.345 0.040
Experience lettuce in 0.053 0.371
Network assets 0.872 0.671
Membership in grower association(s) 0.351 0.009
Number of upstream linkages 0.936 0.278
Number of downstream linkages 0.720 0.451
Environmental assets
Availability of water in the farm 0.351 0.002
Altitude of the farm 0.076 0.130
History of bail damage 2003-2008 -0.234 -0.089
Distance to paved roads 0.351 0.034

 

Instrumental variables: kilometers from house to farm. household size.

"‘ Dropped from regression as explanatory variable.

125

 

Table 4.10 Hausman endogeneity test on membership in association(s)

 

Explanatory variable Coefficient p-value
Constant 0.321 0.078
Physical assets (pa)
Total land worked 0.151 0.057
Land owned 0.819 0.473
Average lettuce plot size 0.234 0.000
Number of cycles produced 0.103 0.000
Owns a vehicle 0.351 0.436
Owns livestock 0.702 0.450
Irrigation equipment for lettuce 0.351 0.443
Human assets
Age of head of household -0.201 0.759
Education of head of household 0.284 0.009
Household size 0.134 0.899
Household members working on farm 0.251 0.444
Off-farm family labor 0.234 0.793
Farm hired labor for lettuce -0.341 0.361
Experience lettuce 0.1 17 0.075
Network assets 0.351
Membership in grower association(s) 0.227 0.023
Residuals 0.297 0.571
Number of upstream linkages 0.523 0.001
Number of downstream linkages 0.223 0.003
Environmental assets
Availability of water in the farm 1.222 0.000
Altitude of the farm 0.281 0.088
History of hail damage 2003-2008 -0.227 0.078
Distance to paved roads -0.049 0.029

 

Instrumental variables: kilometers from house to farm. household size.

126

Table 4.11 Hausman endggneity test on experience growing lettuce

 

Explanatory variable Coefficient p-value
Constant 0.682 0.135
Physical assets (pa)
Total land worked 0.391 0.227
Land owned 0.089 0.122
Average lettuce plot size 0.275 0.000
Number of cycles produced 0.300 0.000
Owns a vehicle 0.421 0.406
Owns livestock 0.257 0.425
Irrigation equipment for lettuce 0.629 0.438
Human assets
Age of head of household -0.418 0.750
Education of head of household 0.327 0.009
Household size 0.466 0.809
Household members working on farm 0.889 0.421
Off-farm family labor 0.239 0.663
Farm hired labor for lettuce -0. 144 0.342
Experience lettuce 0.392 0.055
Residuals 0.234 0.642
Network assets
Membership in grower association 0.234 0.021
Number of upstream linkages 0.351 0.01 I
Number of downstream linkages 0.936 0.013
Environmental assets '
Availability of water in the farm 1.236 0.000
Altitude of the farm 0.234 0.079
History of hail damage 2003-2008 -0.755 0.068
Distance to paved roads -0.1 17 0.022

 

Instrumental variables: kilometers from from house to farm, household size.

127

Table 4.12 Summary of significance between physical assets among HSSCs
and Traditional Suppliers dufingg008, 2004 and 2001
Mean Comparison Sig. (years)

 

 

2008 2004 2001

Number of households (a) 151 176
Land-related assets (Ha)

Land ownership 0.000*** 0.000*** 0000*”

Land rented 0,000’“M 0000*” 0085*

Land share-cropped 0.053 * 0.003 * * 0.102

Total land farmed 0.000*** 0.000*** 0000*”
Land Value (USS'OOO/Ha) 0.136 0.569 0.780
Lettuce area planted(b) 0.0001: * at: 0000* at at: 0.176
Average lettuce plot 0.000* "‘ * 0.000* * * 0039* *
Number of cycles per year 0,000* * * 0.000* * * 0023* *
Cycles*average lot size 0000*“: 0000*:84- 0.008***

 

Mean comparisons between household groups calculated using F-test. *: p < 0.1;

“z p < 0.05; “*2 p < .01

a) Total number of households

b) Lettuce area planted = # of production cycles * average lettuce plot size

128

CHAPTER 5

SUMMARY AND CONCLUSIONS

This dissertation contributes to the development literature in three major areas.
First, it helps to fill a knowledge gap on understanding how market channels evolve over
time using the fresh lettuce case as the target product. Second, it delves further into the
question of why and how some small farmer organizations succeed in accessing more
profitable markets even under constrained access to production factors, information and
networks. And third, it provides newfound evidence on the role of key assets that enable
small farmer organizations to compete in HSSCs sustainably. The guiding context, major
findings, limitations of this study are offered below as well as recommendations for
future research.

The analysis started with a review of the effects of globalization on developed-
country markets and the opportunities to developing countries in providing year-round
availability of mainstream fruits and vegetables. As such, advances in post-harvest
technology enabled countries in the southern hemisphere such as Chile, Argentina, South
Africa and New Zealand to provide those products counter-seasonally to the north,
thereby creating economic growth opportunities for farmers and exporters able to supply
them. Increased demand for diversity, taste and nutrition also led to important
development in consumer behavior towards foreign foods giving origin to the NTAX
market boom. As increased multinationalization of global companies increased during the
19908, change arrived to Central America’s produce sector by enabling large supermarket
and foodservice companies compete freely for larger market shares. With big names in

the industry like Wal-Mart and McDonalds also came standards and tougher market entry

129

conditions to protect their reputation while tapping on the growing profitable markets in
Central-America. This is how HSSCs originated in the Central—American Region and this
dissertation offers a better understanding on what these market channels mean to the
small, resource-constrained farmers in their quest to access opportunities for higher

income and better profits.

The evidence collected through this research points out that the benefits accrued
by small farmers who have successfully entered HSSCs for years have surpassed the
income generation performance of farmers working in traditional markets with a product
such as fresh lettuce. From a general stand point, the capacity of farmers to work in
groups has represented a major advantage characterized as a network assets. F armer
groups that have entered the circle of suppliers to HSSCs soon realized that important
trade-offs have to be made in order to remain competitive, such as working only with the
best resourced farmers.

The organizations working in producing and supplying lettuce have adapted to
change over time. This generated the question of how small farmer organizations reached
a high level of competitiveness in a value chain highly sought after larger suppliers. The
answer has been greatly due to their mixed set of asset endowments and management
structures to avoid the problems that constrained other small farmers historically. One
example of long-term market access performance in HSSCs has been Labradores Mayas.
The entrepreneurial capacity of LM has been remarkable over the past ten years and this
dissertation outlines key aspects of their achieved success. Escaping the fate of several
agricultural c00peratives in developing countries that lost control of their management

capacity to stay profitable, LM provided examples of how a grower entrepreneurial group

130

built among small-scale farmers satisfy the demand of difficult, but more profitable
market channels. Four key organizational strategies were explained as part of the
management formula for LM: (1) Working with growers that minimize costs and increase
group performance. (2) Controlling origin, distribution, and use of key production inputs.
(3) Increasing efficiency through expansion of networks. And (4) benchmarking group

performance against larger groups and companies.

These four strategies are said to be based on emulating private sector competitors,
but LM has truly managed their own evolution despite of the examples and models they
try to follow. What is the future of organizations like this? We may not know that until
more research is generated. However, the organization, management and overall modus
operandi of LM outlined in this study should guide the path to success for similar small

farmer organizations within and outside the conditions of the Guatemalan highlands.

It was established that fresh lettuce is no an easy agricultural product for small—
scale, resource constrained farmers. It is produced, processed and supplied within a
complex value chain with several propensities to failure that range from increased supply
requirements to climatic conditions. While traditional market channels can buy different
quality levels at a price, HSSCs have move beyond traditional market quality levels to
add a number of additional visible (tangible) and invisible but auditable (intangible) set
of standards. As a result, it was established that even if HSSCs represent incentives to
farmers in terms of higher profits and long-term business relationships, it also requires

some capacities in terms of their ability to produce up to standard, organize to compete

131

and follow the dispositions of their management team with no room for opportunitistic
behavior that could risk the reputation of the whole group.

How then LM is still supplying a difficult, highly perishable product successfully?
Part of the answer lied in the asset mix that functioned as determinants of adoption of
HSSCs entry and sustained entrance standards. A development debate has been going on
for over twenty years on similar questions concerning small farmers’participation in such
markets. In this debate, Guatemala has been a particular location to assess how small
farmers’ asset endowments allow them to acquire the inputs, services and connections for
specific market channels. Using some of the seminal research on market channel
adoption, particuarly NTAX, this dissertation embarked in empirical research to answer
this question for the successful case of small farmers in the Guatemalan highlands
producing lettuce. The major objective of the empirical research was to evaluate the role
of small-scale lettuce farmer physical, human, network and environmental assets in
accessing HSSCs.

Building mainly on NTAX research on adoption theory, production functions are
used to represent the decision making process of the small farmer when opting for new
market channels. Through this research it was Ieamed that further attention has to be
placed on the most critical farmer assets such as education and network assets as these
play an important role in allowing the farmer not only to understand the technology
required to satisfy HSSCs standards, but to adjust to changing conditions in such markets.

Finally, this study found no major limitations in scope, research tools and data.
However, the basis gathered in this study on the building of empirical model estimation

for market channel choice adoption should lead further research to the standarization of

132

the methods. Ideally, software could be developed to calculate the critical asset levels of
farmers combined with the target market requirements to produce a likelihood or

probability number that could hint level of farmer or farmer group readiness (e.g., from 0
to l or simply a percentage base qualification) to access such dynamic, but better paying

markets.

133

APPENDIX

Value Chain Study Protocol

General objective of the study

Improve the understanding of the lettuce value chain in Guatemala and the role of small,
land scarce farmers.

Expected outcomes of the study

Generate knowledge on alternative value streams for lettuce producers in
Guatemala

Establish and describe the existing networks in support of farmers and farmer
organizations to access markets

Identify constraints and opportunities related to farmers assets (physical, human
network and environmental) for smallholders producing lettuce

Generate a value stream map identifying the main links of the market structure for
smallholders.

Key features of the value chain study

Collection of secondary data (national agriculture census data, NGO reports, etc.)
Key informant interviews with an open-ended questionnaire addressing
production and marketing aspects

A focus group with industry stakeholders to validate predominant findings about
alternative value streams, constraints and opportunities

134

Table A.l. Research topics and expected number of interviewees and determination
of persons to be interviewed

 

Actors to be Interviewed

 

 

 

 

 

 

- NG
Key topIcs to Supemtarkets/
. , FO Altemative . Private 0
Include durlng Small _ . Foodservrcc _ Government
leade Wholesalers Intermediary Servrcc _ ass
interview producers company . agencres ,
rs buyers provrders ocr
agents
ates
Productivity J J J J
Technology J J J J J J J
Costs J J J J
Income / on J J J J J J J J
and off farm
employment
Organizational J J J J
structure of the J

farmers’
organizations

Organizational J J J J J J J J

functions of the

 

farmers‘
organizations

Contract terms J J J J J J J

/ obligations

Chain J J J J J J J

governance /

 

 

decision-
making
Compliance J J J J J J J

monitoring /

 

sanctions

 

Partner access J J J J
to support J J

services

 

Number of

interviewees

 

 

 

 

 

 

 

 

 

 

 

135

Open-ended Interview Guide

I Production Level

1)
2)
3)
4)
5)
6)

7)
3)

9)

Why did you decide to grow lettuce?

Why do you think this area or your farm is good to produce lettuce?

How well have you done in your harvests in the past few years?

How do you source your seeds, fertilizers and other inputs?

Do you work on your own or do you produce together with family and/or friends?
Do you belong, or have belonged to farmer organizations? If so, what is the
assistance you have received from this/these association(s)?

What is your major concern regarding production aspects of lettuce?

Can you tell me in what stage of the production you know if the harvest will be
good or not?

What would you say are your most worrying technical factors in producing
lettuce?

10) Have you improved your technology to produce lettuce in the last few years?
11) Who or what is your major source of support in the production of lettuce?

I Commercialization Level
12) Where do you sell your lettuce?
13) Why have you chosen this market?
14) Have you supplied to other markets in the last five years?
15) What is your major concern in marketing your lettuce?
16) What is your major source of support to market your lettuce?
17) Have markets improved for your lettuce? If yes or no, how?

Preliminary data analysis

Transcription of field notes
Preliminary value chain map
Linking answers to research questions

Secondary interview visits

Final data analysis plan

Data coding

Identification of concepts and themes

Revision of value chain map

Mapping of value streams

Listing concepts and themes (preliminary data packing)

Focus group

Present preliminary concepts and themes to stakeholders

136

0 Moderate discussions on challenges and opportunities identified
' Validate concepts and themes identified
' Refine concepts and themes to link evidence

Final report

' Drafi preliminary draft

137

 

Case Study Protocol

General objective of the study

To determine what is going on with the farmer organization(s) under study in order to test
the theory on the role of network assets in facilitating small farmers’ access to HSSC’s.

Purpose of the study
To answer the following research questions:
0 How has LM arrived to this level of market access performance in HSSCS?
o How can the LM organization model be characterized for other farmer
organizations to pursue access to HSSCs?
Key question areas during interviews.
(1) Please tell me about organization’s history, development and growth
(2) Please share with me any analysis on the company's internal strengths and weaknesses

(3) What you consider has allowed your organization to work well over the years

(enabling environment)

(4) Have you done a SWOT analysis? When, what was the result?

(5) Is there a long-term plan?

(6) How do you plan to increase your market share with your current clients?
(7) How are decisions made within the organization?

138

Table A.2 Summary of research issues and propositions

 

 

Research . . Operationalization of
. Proposrtlons . .
Questions proposrtlons
How have Proposition 1: Barham and Chitemi (2008)

network assets
(social capital,
collective
action and
network
capital)
supported
farmers move
from

Farmers in
Labradores Mayas
exhibit more
interpersonal trust
than farmers
outside of
Labradores Mayas

formulated the following questions
to interviewees( 1) most members
in your group can be trusted
(General Trust); (2) most members
in your group are willing to help if
you need it (Help Trust); and (3) in
your group, members can
generally trust each other in
matters of lending and borrowing

 

 

network assets
(social capital,
collective
action and
network
capital)?

traditional money (Money Trust). Following
market to a similar codification approach,
HSSC’s? answers were coded from 1 to 3 by
group.

. How are Proposition 2: Wealth heterogeneity will be
Labradores Farmers in measured using household
Mayas farmers Labradores Mayas ownership of physical and
different from exhibit higher financial assets such as land,
traditional wealth livestock and vehicle ownership.
market heterogeneity than Since this case study will
suppliers in farmers in other interview leaders and founders of
respect to groups the organizations, the information

to be gathered will be perception-
based on a wealth ranking scale
from 1 to 3, 1 being the lowest
level of wealth.

 

Proposition 3:
Labradores Mayas
exhibit lower levels
of poverty than the
other non
successful
organized group.

Operationalized by comparing the
number of members with different
wealth rankings provided across
groups from very poor, poor,
average, rich and very rich. Land
ownership and other assets
associated with these rankings will
be developed based on findings
from the value chain study.

 

 

Proposition 4:
Labradores Mayas
exhibit more past
successful
experiences than
the other non
successful
organized group.

 

This will be measured in years of
successful years accessing HSSC’s
or other market venues. This will
still be a perception-based
approach expected to provide
robust comparison data with the
non successful group.

 

139

 

Table A.2 Continued

 

3.

Proposition 5:
Labradores Mayas
exhibit more
appropriate
leadership than the
other non successful
organized group.

This will be measured in terms
of group norms to choose their
leaders and a perception based
on the effectiveness of past
leaders of the organization.

 

Proposition 6:
Labradores Mayas
exhibit more group
activities than the
other non successful
organized group.

Group activities are categorized
as services or activities provided
to the member base such as
credit programs, collective
marketing, input purchasing in
bulk, labor-sharing activities,
collective seedling nurseries and
acquisition of services such as
pest control advisors and
demonstration plot hands on
training

 

4. How are other
asset
endowments
(physical,
human and
environmental
assets) related
to network
assets?

 

 

Proposition 7:
Labradores Mayas
exhibit more market
linkages downstream
in the value chain to
wholesalers and
retailers than the
other non successful
organized group.

 

With the assistance of the value
stream map to be developed in
the value chain study (essay 1)
the number of effective linkages
measured in terms of existing
market channels will be
measured for each group.

 

140

 

Survey instrument

The role of household assets in small farmer access to high-standards supply
chains

Verbal Consent Form

Participant Name:

 

Interviewer Name:

 

Date: Time

 

 

Hello, my name is Luis Flores. I am a graduate student at Michigan State University, in
the United States. As part of a Michigan State University study, we wish to learn about
how lettuce farmers make decisions about to whom they supply their produce.

The interview-discussion should take no more than about 45 minutes. Your participation
in this research interview is voluntary and you may refuse to answer any or all questions.
Furthermore, you can stop the research interview at anytime without any problems or
negative consequences. It is important for you to know that there are no right or wrong
answers.

Your identity and your responses will be kept confidential and your privacy will be
protected to the maximum extent allowable by law. While the information collected
during the interview may be used in reports and publications, your identity will not be
revealed with only a code number used to identify the source of the data.

As a participant in this study, I will be audio taping the interview to ensure that I do not
miss anything we talk about. All information I collect during the interview, including any
audio-recordings, will be stored on secure, password-protected computers and destroyed
as soon as practicable.

f you have questions or concerns about your role and rights as a research participant, or
would like to register a complaint about this study, you may contact, anonymously if you
wish, the Michigan State University's Human Research Protection Program at 517-355-
2180, Fax 517-432-4503, or e-mail irb@msu.edu or regular mail at 202 Olds Hall, MSU,
East Lansing, MI 48824.

Do you have any questions? Yes No [if yes, answer questions and proceed]

May I begin? Yes No [If no, thank and end]

141

 

Information sheet

If you have any questions about this study, you may contact any of the following people:

Study Leaders:

Michael Kaplowitz, J .D., Ph.D.

Assoc. Prof, Environmental Law & Policy
305C Natural Resources

Michigan State University

East Lansing, MI 48824, USA

T: 001 517-355-0101 F: 001 517-353-8994
E: kaplowit@msu.edu

Research Assistant:

Luis G. Flores

1203 Westmoreland Ave
Lansing, MI 48915

Tel: 00-1-517-432 2214
E-mail: floreslg@msu.edu

Small lettuce farmer’s household physical, human, network and environmental
assets as explanatory variables in farmers access to HSSC’s versus traditional
markets

Survey Questionnaire to Small Lettuce Producers

Part I. General Information

Enumerator: Fill this part before the interview

 

 

 

 

 

 

Date: Interviewee name:

Time: Market channel:

Enumerator code: Municipality: Village:
How twfi there:

GPS coordinates (to be filled out by Supervisor:

 

 

 

Enumerator: if the answer is no. discontinue survey here.
Enumerator: Please use the following codes for questions without an answer:

N0 = I

Doesn ’t know, doesn ’I respond = 2

Other: 3

Sr. Madame,

My name is . I would like to talk to you for some 30 minutes. I’m working for Mr. Luis Flores,

he is conducting a research project to evaluate the characteristics of your farm to produce lettuce. The study
is being done in this area of Guatemala and is part of the requirements of Mr. Flores for his graduation in
the department of Community, Agriculture, Resource and Recreation in Michigan State University. I am
here with you today because I have been informed that you produce lettuce for (check one or both if
applies) the _ (0) traditional __ (l) supermarket channel. 15 this correct? Yes No

142

 

Would jou kindly provide me...

 

Your name:

 

Do you have a home phone number?

 

Do you have a cell phone number?

 

Does your farm have a name?

 

 

 

Part II. About the farm, physical and environmental assets

Enumerator: formulate the question and gather data in the dimensions that are easier for the farmer.
Calculate total in manzanas and hectares at the end of the interview

I C uerda= 0.14 manzanas = 1000 m2

I Cuerda= 0.] Ha

I manzana = 0. 7 Ha.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1. Land ownership and management (A) (B)
This agriculture 5 years ago
ear (2007-2008) (2003-2004)
1.1. How many manzanas do you own in this farm and other Manzanas_ Manzanas___
farms that you manage? nsnr nsnr
1.2. How many manzanas do you rent from others? Manzanas_ Manzanas_
nsnr nsnr
1.3. How many manzanas do you rent to others outside your
household? Manzanas_ Manzanas_
nsnr nsnr
1.4. How many manzanas do you work under share cropping Manzanas_ Manzanas_—
outside your household? nsnr nsnr
Enumerator, calculate total land owned and managed
. Do you know the value of a manzana in this area? nsnr
. What is the cost of renting of a manzana in this area? nsnr
(A) (B)
This agriculture 5 years ago
year (2007-2008) (2003-2004)
4. How many manzanas of lettuce did you grow? Manzanas_ Manzanas_'
nsnr nsnr
5. How many cycles of lettuce do you grow per year? Manzanas_ Manzanas_
Enumerator: A cycle in the area is takes 70 days from nsnr nsnr
lanting to harvesting.
6. What is the usual size of your plots?
7. In the calendar year, in what month do you start planting Manzanas_ Manzanas_
and in what month do you finish harvesting? nsnr nsnr
Enumerator, calculate total lettuce land cropped
8. What other crops did you grow Manzanas (C)
five years ago? (2003-2004) (A) (B) "5'"
Other Basic grains
vegetables
8.1
8.2

 

 

143

 

 

 

 

9. What other crops have you
grown this year? (2007-2008)

Manzanas

 

(A)
Other

vegetables

(3)

Basic grains

(C)

“Sill"

 

9.1

 

9.2

 

9.3

 

9.4

 

9.5

 

9.6

 

9.7

 

 

9.8

 

 

 

 

 

10. Do you have irrigation in your farm? Yes

If yes, proceed to 12-13.

No

 

This agriculture year (2007-2008)

 

What type of irrigation

(A)
Lettuce (M2)

(3)

Other vegetables (Mz)

(C)
Basic grains
(M2)

 

10.1 Furrow, gravity

 

10.2 Drip

 

10.3 Overhead

 

10.4No irrigation (please
indicate manzanas)

 

 

 

 

 

 

Five years ago (2003-2004)

 

What type of irrigation

(A)
Lettuce (Mz)

(B)

Other vegetables (M2)

(C)
Basic grains (Mz)

 

10.5 Furrow, gravity

 

10.6 Drip

 

10.7 Overhead

 

 

10.8 No irrigation
(please indicate
manzanas)

 

 

 

 

 

1 1. What is the source of your irrigation water?

1 1.1 (A) river (B) How far is it from your farm?

1 1.2 (A)Artisanal well

11.4 nsnr
11.4 Municipal?

 

(B) how many
1 1.3 (A) Deep well (B) How many C) How far is the water table? Meters

12. Do you have access to potable water at the farm?

Yes No

144

Meters

"T"

If yes,

12.1 Is it municipal?

12.3 Is it privately owned?
12.4 Other

13. In the past five years, has frost or hail damage a problem in your farm? Yes
If yes, mark the years and months

13.1 2008 _ Month:
13.1 2007 __ Month:
13.1 2006 _ Month:
13.1 2005 __ Month:
13.1 2004 __ Month:

14. Do you own livestock? Yes No
Ifyes, proceed with 14.1-14.2

 

No

 

(A)
This agriculture
year (2007-2008)

(B)
5 years ago
(2003-2004)

 

14.1 Heads of bovine livestock heads
nsnr

heads
nsnr

 

14.2 Heads of equine livestock heads
nsnr

 

 

heads
nsnr

 

 

15. Do you own a truck or a vehicle? Yes No
If es, proceed with 15.1-15.2

 

15.1 When did you get it? Month

Year

 

15.2 Make Model

 

 

Year

 

 

16. What is the distance from your farm to:
16.1 The nearest paved road Km
16.2 What is the name of the place where you sell your products?

 

 

16.3 How far are you from this sale point Km

 

16.4 What is the name of the community where you source your inputs?
16.5 How are is the farm from this community? Km

Part III. About human and network asset-related areas
17. May I ask how old are you?

18. How many years of school did you complete?

19. How many people live in your household?

20. Do you belong to an association?

21. How many people work with you at the farm including you?
Family labor

 

 

# 2003-2004

# 2007-2008

 

Feminine

 

 

 

Masculine

 

 

145

 

 

 

22. How many months/days per year do you work for someone else?

22.1. For how many Months Days

23. How many months/days per year are members of your household working for someone else?
23.1. For how many Months Days
24. How many people that are not from your household work for you? __
24.1. For how many Months Days

 

25. When did you start producing lettuce for the first time?

26. Number of input suppliers from whom you source:
26.1. Production inputs
26.2. Financial credit
26.3. Training
26.4. Packaging material
26.5. Other

 

 

 

 

27. Number of buyers with whom do you engage in transactions (to whom you sell)?
27.1 Supermarkets
27.2 Wholesalers
27.3 Traditional Markets
27.4 Exporters
27.5 Others

 

 

 

 

 

28. Sources of information about price?
Buyer[ ] Trader[ ] Market[ ]Other[ ]

 

29. Number of buyers you discuss prices before selling?
29.1. Supermarkets
29.2. Wholesalers

29.3. Traditional Markets
29.4. Exporters

 

 

 

 

 

 

29.5. Others

30. Time knowing the buyer?

30.1. Supermarkets Years [#__] Months [#_ ] Never Know [ ] Other [ ]
30.2. Wholesalers Years [#_] Months [#_ 1 Never Know [ ] Other [ ]

30.3. Traditional Markets Years [#_] Months [#__ ] Never Know [ ] Other [ ]

30.4. Exporters Years [#___] Months [#__ ] Never Know [ ] Other [ ]

30.5. Others Years [#_] Months [#__ ] Never Know [ ] Other [ ]
31. How do you communicate with your buyers?

31.2 Supermarkets Phone [ ]; Fax [ ] ; Radio [ ] ; Field visits I l; lntemet [ ]

31.3 Wholesalers Phone [ ]: Fax [ ] ; Radio [ ] ; Field visits [ ]; lntemet I 1

31.4 Traditional Markets Phone [ ]; Fax [ ] ; Radio [ ] ; Field visits [ ]; lntemet [ ]

31.5 Exporters Phone [ ]; Fax [ ] ; Radio [ ] ; Field visits 1 ]; lntemet l I

31.6 Others Phone [ ]; Fax [ ] ; Radio [ ] ; Field visits [ ]; lntemet [ ]

32. Where do you deliver your products?

32.1 Supermarkets At Farm [ ]; Buyer pick up [ ] ; Deliver to Buyer [ ] ; Other[ ]____
32.2 Wholesalers At Farm [ ]; Buyer pick up [ l ; Deliver to Buyer [ ] ; Other [ ]_____

32.3 Trad. Markets At Farm [ ]; Buyer pick up [ ] ; Deliver to Buyer [ ] ; Other[ ]_____
32.4 Exporters At Farm [ ]; Buyer pick up [ l ; Deliver to Buyer [ ] ; Other[ ]______

32.5 Others At Farm 1 ]; Buyer pick up [ ] ; Deliver to Buyer[ ] ; Other [ ]_____

 

146

Enumerator:
1) Develop a map of the farm location and take GPS reading.
2) Go back to questions that were not completed up to satisfaction
3) Ask interviewee if he/she has any questions
4) Ask interviewee if you could come back later on in case you have further questions
5) Thank farmer kindly

147

   

M'1111111111111111lllllllllfiltjllllllf