, A 3a...
ll . f
.: .tnfltiifllzfivvfd.i
’i'.‘ 1 n It!
. 3.14143...
! . in}! ,
v-39,“ a:

.1.

 

3:
.n 4.. u. i...

1. «t
unfihm
annfiw .

A .3

t .2
vxhsun... .
3 I...‘

 

 

 

.e .
-3“.

3&2..$&»u..
...,v ...n {f .
Innsxflnr‘

“narramu?

. nu: ! 5......
:zqg
;:nu

0

ll ”4.“. «1...

.s .3 1 i.

‘ 2133...?

2,. I. . , .2113:

-, iiizeii ,
75‘ ‘1‘! .

 

 

 

 

 

  

 

 

, : . A ......' 51%?“ u
‘ 7 ‘ . . V .3 , iv .
damage, 3.. uh? 3%.,i...... .A

:3. . .

  

 

 

 

71-...
-§...H£r.~
3 A
fin? v.3
.r.

lots-X 4.).) 1.25:... »
.: ... AK... .
.. . ..
r», ... t f
, . 5.}. .

4.x? pawn?»

  

71126919
LIBRARY
7% Michigan State
University

This is to certify that the
thesis entitled

CONSTRAINTS TO AND OPPORTUNITIES FOR
DEVELOPING A BEAN SEED PRODUCTION AND
MARKETING SYSTEM IN HONDURAS

presented by

Wolfgang Pejuan

has been accepted towards fulfillment
of the requirements for the

Master of degree in Agricultural Economics
Science

 

 

72M WEE/ML

Major Professor’s Signature
I 2'] l3! 0 5

Date

 

MSU is an Afi‘innative Action/Equal Opportunity Institution

-.-.a:-._n------.---c-.---.-.-----

‘-§-n-l-l-I-C-‘-l-t

 

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

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 cICIRC/DateDueinddp. 15

CONSTRAINTS TO AND OPPORTUNITIES FOR DEVELOPING A BEAN
SEED PRODUCTION AND MARKETING SYSTEM IN HONDURAS

by

Wolfgang Pejuan

A THESIS

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

MASTER OF SCIENCE

Department of Agricultural Economics

2005

ABSTRACT

CONSTRAINTS TO AND OPPORTUNITIES FOR DEVELOPING A BEAN SEED
PRODUCTION AND MARKETING SYSTEM IN HONDURAS

By
Wolfgang Pejuan
Recent studies have found that over 50% of bean farmers in the main bean-
producing areas of Honduras planted modern varieties. However, many of the farmers
planted varieties released up to 10 years ago and most farmers obtained seed fiom
neighbors or farmer-saved seed. This indicates a need to increase physical supply and
expand farmer access to modern bean varieties. One option for meeting this need is to
establish a seed system in which small-scale farmers are contracted to multiply seed of
modern varieties under the supervision of Zamorano and market seed through an input
supply firm. This study looks at the feasibility of implementing such a scheme. A rapid
appraisal was conducted to collect information from key informants to describe the bean
seed subsector in Honduras and identify potential small-scale bean seed producers and
marketing agents. In addition, 72 farmers in El Paraiso and Olancho were surveyed to
assess the demand for improved bean seed. Amadeus 77 was the variety with the highest
demand. Bean seed production budgets were constructed and used to determine if it was
feasible to market improved bean seed at department capitals and smaller towns for L
12/1b (L 26.4/kg). At a price of L 12/lb (L 26.4/kg), the estimated seed demand for
Amadeus 77 in El Paraiso and Olancho is of approximately 24 mt and 54 mt. At this
price and demand level, the projected demand is sufficient to motivate input marketing

firms to distribute the seed produced under contract by small-scale seed producers.

C0pyright by
Wolfgang Pejuan

2005

Dedication

I dedicate this thesis to my wife, Laila, for all her love, encouragement, and

support. Thank you for being my motivation and simply for being who you are.

Also, I would like to dedicate it to my mother, Ilma, who has always supported me

in my career and life.

iv

AKNOWLEDGMENTS

I would like to express my special gratitude to my major professor and thesis advisor,
Dr. Richard Bemsten, for giving me the opportunity to come to MSU and for his invaluable

teachings. Also, for his guidance, support, and effort devoted to revise my drafis.

I would also like to thank my committee members, Dr. Irvin Widders, Dr. Mywish
Maredia, and Dr. Stephen Harsh, for the helpful contributions.

To Dr. Scott Swinton, Dr. Roy Black, Dr. Robert Myers, and Dr. Peter Schmidt for

contributing on my research.

To the BEAN/COWPEA Collaborative Research Support Program for funding my
studies at MSU and research in Honduras. Also, the Bean/Cowpea representatives in Central

America that helped me out during my research: Juan Carlos Rosas and his staff.

To the Fulbright Program for sponsoring my Master’s degree studies.

To my brother José Alberto, and to Danilo Escoto, Reinaldo Arguefial, and Jorge
Sénchez for helping me carry out my survey.

To my family in law who have welcomed me and my family into their family.

Thanks to my sister, Sayda, for hearing me out when I have needed support.

Table of Contents

List of Tables .......................................................... _x
List of Figures ........................................................ xfi
1. Introduction .......................................................... l
1.1 Objective of Study .............................................. 3

1.2 Research Questions ............................................. 5

1.2.1 Demand for Modern Bean Varieties ........................ 5

1.2.1.1 Farmer ........................................ 5

1.2.1.2 Bean Traders ................................... 5

1.2.2 Supply of Modern Bean Varieties ........................... 6

1.2.2.1 Research Agencies ............................... 6

1.2.2.2 Bean Seed Growers (Previous) ..................... 6

1.2.2.3 Bean Seed Growers (Potential) ..................... 1

1.2.2.4 Government Agencies ............................ §

1.2.2.5 Non-Govemment Organizations .................... g

1.2.2.6 Private Sector Bean Seed Firms ..................... 2

1.2.2.7 Input Marketing Firms ........................... _19

1.2.2.8 Bank Officials ................................. m

1.3 Thesis Outline ................................................ _1_1

2. Conceptual Framework ................................................ _1_2
2.1 Overview .................................................... E

2.2 Subsector Analysis ............................................ L;

2.3 Seed Systems ................................................. _13

2.3.1 Stages of Seed Development .............................. Q

2.3.2 Types of seed systems ................................... ,1_5

2.3.3 Informal Seed Systems in Central America .................. _1_6_

2.3.4 Informal Seed Systems in Honduras ........................ _1_7

2.4 Function of Seed .............................................. _1_8_

2.5 Seed Demand ................................................. 12

2.5.1 Seed Rate ............................................ _2_O

2.5.2 Area Planted .......................................... 2_l

2.5.3 Varietal Choice ........................................ _2_l

2.5.4 Hedonic Price Model . .................................. 21

2.5.5 Willingness to Pay ..................................... _25

2.6 Seed Supply .................................................. 26

2.6.1 Supply function ........................................ _2__6_

2.6.2 Budgeting ............................................ 27

2.7 Transaction Costs .............................................. 23

2.7.1 The Nature of Transaction Costs .......................... 2§

2.7.2 Farmers’ Transaction Costs .............................. _2_9

vi

2.7.3 Seed Firms’ Transaction Costs ............................ fl

2.7.4 Govemment’s Role in Reducing Transaction Costs ............ 3_1

2.8 Summary .................................................... 3_1
3. Research Methods .................................................... fl
3.1 Data Collection ............................................... 3_4
3.1.1 Rapid Appraisal ....................................... 33

3.1.1.1 Input Dealers. .................................. fl

3.1.1.2 Seed Firms ................................... 15

3.1.1.3 NGOs and Projects ............................. 15

3.1.1.4 Banks ........................................ &

3.1.1.5 Research Agencies ............................. 16

3.1.1.6 Seed Certification Agency ....................... 31

3.1.1.7 Farmer Groups ................................ 12

3.1.1.8 Traders ...................................... 3_8

3.1.1.9 Budget Data Collection .......................... 3_8_

3.1.2 Farmer Survey ......................................... _3_9_

3.1.2.1 Questionnaire .................................. 43

3.2 Empirical Methods ............................................. $4;
3.2.1 Empirical Model for Demand ............................. fl

3.3 Empirical Model for Seed Supply ................................. 15
3.4 Summary .................................................... fl
4. Bean Seed Supply in Honduras .......................................... 41
4.] Seed Production Stages ......................................... 41
4.2 Seed Supply Agents ........................................... fl
4.2.1 Seed Certification Service ................................ 48

4.2.2 Breeder, Foundation, and Registered Bean Seed Producers ...... 51

4.2.2.1 Zamorano/PIP ................................. 5;

4.2.2.1.] Production ............................. 51

4.2.2.1.2 Quality Control ......................... fl

4.2.2.1.3 Distribution and Sales .................... 53

4.2.2.1.4 Prices ................................. ,5_3_

4.2.2.2 DICTA ....................................... 51

4.2.2.2.] Production ............................. i3

4.2.2.2.2 Quality Control ......................... fl

4.2.2.2.3 Distribution and Sales .................... 14

4.2.3 Certified and Commercial Bean Seed Producers .............. _5_§

4.2.3.1 Zamorano/Zamoempresa de Cultivos Extensivos ...... §_5_

4. 2. 3. 1. 1 Production ............................. _5_§

4.2.3.1.2 Quality ................................ £5

4.2.3.1.3 Sales ................................. 56

4.2.3.1.4 Contract Terms ......................... 56

4.2.3.1.5 Contractee’s concerns .................... 58

vii

4.2.3.2 Hondugenet ................................... 82

4.2.3.2.] Seed Production ........................ 62

4.2.3.2.2 Contract terms .......................... 66

4.2.3.2.3 Sales ................................. 61

4.2.3.2.4 Hondugenet Bean Seed Production Role ..... 61

4.2.4 NGOs and Artisan Seed Producers ......................... 62

4.2.4.1 NGOs ........................................ 62

4.2.4.2 Role of CIAT .................................. 68

4.2.4.3 ARSAGRO ................................... _6_3_

4.2.4.4 Potential Bean Seed Producers .................... 6_6

4.2.5 Input Distributor ....................................... 66

4.2.5.1 DUWEST ..................................... 6_6

4.2.6 Agricultural Banks ..................................... 61

4.2.6.1 Recent Lending Activities ........................ 68

4.2.6.2 Banking Operations ............................. 68

4.2.6.3 Loan Conditions ................................ 62

4.2.6.3.1 Interest Rate ........................... 62

4.2.6.3.2 Time Horizon .......................... 62

4.2.6.3.3 Collateral .............................. 7_0

4.2.6.3.4 Group Loans ........................... ZQ

4.2.6.3.5 Interest in Loaning to Scheme Farmers ....... 11

4.3 Summary .................................................... Z1
5. Bean Seed Demand In Honduras .................................... ' . . . . E
5.1. Farmer’s Characteristics, Seed Use, Production, and Sales ............. 14
5.1.1 Socioeconomic Characteristics of the Farmers ............... B

5.1.2 Bean Area ............................................ B

5.1.3 Source of Bean Seed ................................... 16

5.1.3.1 Reasons ...................................... _71

5.1.3.2 Prices Paid .................................... 7_8

5.1.4 Varieties Planted ...................................... B

5.1.5 Traditional Varieties v.s. Modern Varieties .................. 86

5.1.6 Total Production ....................................... 81

5.1.7 Bean Sales ........................................... 8_2

5.1.8 Prices Received ....................................... 84

5.2. Demand for Modern Varieties ................................... 86
5.2.1 Preferred Bag Size ..................................... 86

5.2.2 Purchase Frequency .................................... &

5.2.3 Price Discounts for Modern Varieties ....................... &

5.2.4 Farmers Preferences for Recent Modern Varieties ............. fl

5.2.5 Factors Associated with Farmer Demand .................... 26

5.2.6 Results ............................................... _91

viii

5.3 Summary .................................................... 22

N ,

6. Bean Seed Production Scheme ........................................ 1(1_
6.1 Supervision of the Bean Seed Scheme ............................. _1_8

6.2 Seed Production Sites and Participants ............................ _162

6.3 Contracts ................................................... 168

6.4 Credit ...................................................... M

6.5 Commercial Seed/Variety ...................................... 166

6.6 Farmer Training and Supervision ................................ 166

6.7 Seed Processing/Packaging ..................................... _166

6.8 Seed Inspection .............................................. _16Z

6.9 Distribution ................................................. 1(fl

6.9.1 Price ............................................... M

7. Summary, Conclusions, and Policy Implications .......................... fl
7.1 Summary ................................................... L

7.2 Conclusions ................................................. 11_

7.3 Policy Implications ........................................... 11__

7.4 Limitations .................................................. L

7.5 Future Research .............................................. 11__
Appendix A. Explanatory variables correlation matrix ........................ _12_
Appendix B. Table of probit predicted outcomes ............................. 12__

Appendix C. Probit and truncated model coefficients of bean attributes and farmers’
characteristics on purchase decision and WTP. ......................... l 1

—d
N

References ...........................................................

ix

List of Tables

Table 1. Number of farmers surveyed by department, region, and village in Honduras,

Bean Survey, 2002. ............................................... 41
Table 2. Steps seed industry producers and processors must follow to obtain seed

certification from CERTISEM, Honduras, 2002. ........................ 42
Table 3. Quality standards for certified and commercial seed, Honduras, 2002. ...... fl
Table 4. Field tolerances for different categories of bean seed, Honduras, 2002. ..... ,5_1
Table 5. Percentage of seed rejection of seed delivered to the Zamoempresa de Cultivos

Extensivos by a representative contractee, Honduras, 2002 ................ 3
Table 6. Seed Production and marketing in Honduras, 2002 ...................... 6_5

Table 7. Value of loans extended to bean farmers by several banks in Honduras, 2002. 68
Table 8. Average area and average of farmers planting com and bean in postrera (2001)
and primera (2002), by department, Honduras. .......................... _76
Table 9. Hectares farmers planted (percentage) to bean by seed sources and department,
during four seasons (postrera 2000 to primera 2002), Honduras. ............ 16
Table 10. Percentage of farmers quoting different reasons for not ever buying certified
seed, by department, Honduras, 2002. ................................. 11
Table 11. Average hypothetical prices (Lempiras/Kg) for bean seed purchases among
neighboring farmers, by department and type of seed, primera 2002, Honduras. 18
Table 12. Area planted (ha) and number of farmers who planted the variety during any of
four seasons, (postrera 2000 to primera 2002), Honduras. ................. 86
Table 13. Percentage of farmers who planted modern varieties, traditional varieties, and

both traditional and modern varieties (not necessarily in the same season) during

four seasons (postrera 2000 to primera 2002), by department, Honduras. ..... 8_1
Table 14. Farmers total production and area by type of variety, postrera 2000 to postrera
2001, by department, Honduras. ..................................... 82
Table 15. Farmers (percentage) preferred use of a variety (or varieties), postrera 2000 to
primera 2002, by department, Honduras ................................ 88
Table 16. Farmers’ sales (percentage) of the 2001 postrera harvest, by type of variety
and department, Honduras. ......................................... jfl
Table 17. Price (Lempiras/Kg) at which farmers sold their postrera 2001 harvest, by type
of variety and department, Honduras. ................................. 86
Table 18. Summary of bean prices (in Lempiras /kg), Honduras, 2002 ............. fl
Table 19. Traders (N = 5) procurement prices of modern varieties in Mercado Las
Americas, July 2002, Tegucigalpa, Honduras. .......................... 81
Table 20. Ranking of six recently released varieties, by department, July 2002,
Honduras. ...................................................... 88
Table 21. Estimated annual demand (kg) for seed of Amadeus 77 at various prices, by
department, Honduras, 2002. ........................................ %

Table 22. List of explanatory variables in determining desirability to buy (DB) and
willingness to pay (WTP) and the hypothesized relationship between these
explanatory variables and farmers’ decision to buy and WTP. .............. fl

Table 23. Marginal effects of bean attributes and farmer’s characteristics on purchase

decision and willingness to pay (WTP), Honduras, 2002. .................. 28
Table 24. Estimated commercial seed production costs in El Paraiso and Atlantida,

Honduras, 2002 . ................................................ fl
Table 25. Seed production training and inspection costs of farmers fields, Honduras,

2002 ........................................................... 110
Table 26. Estimated prices for commercial seed by stage in proposed bean seed scheme, ‘

Honduras, 2002 ................................................. _1_11

xi

List of Figures
Figure]. Factor demand function for single-variable factor .....................
Figure 2. Bean price factor dependence. ....................................
Figure 3. Short-run firms’ supply curve in a competitive market. ................
Figure 4. The map of Honduras ..........................................
Figure 5. Department of El Paraiso, Honduras. ..............................
Figure 6. Department of Olancho, Honduras ................................
Figure 7. Bean seed development stages ....................................

Figure 8. F arrners’ willingness to pay for bean varieties in El Paraiso, 2002, Honduras. 90

Figure 9. Farmers’ willingness to pay for bean varieties in Olancho, 2002, Honduras.

Figure 10. Sample farmers annual demand by variety in El Paraiso, 2002, Honduras. .
Figure 11. Sample farmers annual demand by variety in Olancho, 2002, Honduras. . .

Figure 12 . Proposed bean seed production scheme ...........................

xii

VFW
\1-50

B

41
Al
.47

l

mama

Chapter I
1. Introduction

Dry beans, the second most important grain in Honduras, accounts for an
important part of the basic diet of most of the Honduran population. Currently, bean
productivity is low. Many factors contribute to low productivity, including poor soils,
climate-related factors like drought and hurricanes, farmers’ low levels of input
application, and their limited adoption of recommended agricultural practices, especially
recently-released modern bean varieties which are a key technology that has the potential
to increase farmers’ yields.

There are several reasons why farmers purchase seed. Farmer demand for new
seed can arise from their desire to replace retained seed of an existing variety that is no
longer genetically pure or to obtain a new variety that is higher yielding or is more pest or
disease resistance (Heisey and Brennan, 1991; Tripp, 1997). Also, farmers in developing
countries are sometimes forced to search for seed outside their farm because, due to
poverty, they are unable to harvest and save sufficient seed for the following season
(Tripp, 1997).

Farmer demand for seed of modern varieties (MVs) depends on the performance
of the varieties in farmers’ fields (relative to traditional varieties, TVs), the acceptability
of these varieties in the market, their local availability, and their affordability (Bemsten
and Mainville, 1999). In Honduras, new varieties have been released with high yield
potential and resistance to various diseases, especially Bean Golden Yellow Mosaic

Virus (BGYMV). On-farm trials in Honduras indicated that recently-released MVs like

Tio Canela produce higher yields than traditional varieties (Rodriguez and Viana, 1998).
A recent study found that over 50% of bean farmers in the main bean-producing areas of
Honduras (El Paraiso, Olancho and Francisco Morazan) planted MVs, which indicates
that they see advantages to planting MVs—primarily due to their higher yield potential and
disease resistance (Mather et. a1. , 2002). However, many of the farmers planted varieties
that were released in 1990 and virtually all of the farmers obtained their seed from a
neighbor or planted farmer-saved seed (Mather et. al, 2002).

Farmers’ willingness to purchase an input such as certified seed depends on its
price, their economic situation, and the availability of credit and government intervention
such as subsidies. Given that few poor farmers in Honduras have access to credit, the
relatively high cost of certified seed is a major constraint to its use. For exarriple, in 2002
in Honduras, certified bean seed sold for approximately USS 96/qq (quintal = 100.1bs) ($
2.1 l/kg). In contrast, farmers can typically buy “saved” bean seed from their neighbors
for only USS 24 /qq ($ 0.53/kg) (i. e. 25% of the price of certified seed). Furthermore,
compared to other countries in the region, certified seed is relatively expensive in
Honduras. For example, the price of certified seed in Nicaragua (2002) was US$ 58/qq (S
1.28/kg), while saved seed was valued at US$ 25/qq ($ 0.55/kg).

Utilizing Shaffer’s (1973) characterization of activities governing a food sector,
bean seed systems can be characterized by the agents involved and the coordinating
mechanisms which bring modern varieties to the market. The agents involved in seed
production may include private or public research agencies that develop modern varieties

and produce foundation seed; private seed firms, agricultural schools, and NGOs which

multiply foundation seed to produce certified or commercial seed; a public or private seed
certification unit that verifies the quality of the seed; public or private input marketing
firms that distribute the seed; and other agents like banks and the government that
facilitate bean seed trade by providing loans to farmers and establishing policies that
support seed production and distribution.

As the consumer of the seed, farmers demand specific characteristics in the
varieties that they plant. However, their varietal preferences are strongly influenced by
bean traders, who take into account the preferences of consumers when setting the price
that they will pay for the grain. Consequently, in order to recommend actions to increase
the productivity of a seed system, it is necessary to take into account the varietal
characteristics preferred by farmers, traders, and consumers.

1.1 Objective of Study

The general objective of the study is to assess the feasibility of both increasing the
physical supply of modern bean seed varieties and expanding farmer access to these
varieties. In the proposed seed system model, small farmers would be contracted to
multiply seed of modern bean varieties, under the supervision of Zamorano. Rather than
producing “certified’ seed, the farmers’ seed would be marketed as “Zamorano
Supervised” seed. This seed would be sold at a price lower than “certified” seed, thereby
making it more affordable to small farmers. Regarding marketing, the seed would be sold
in small size bags (size to be determined) and distributed by one or more private sector
firms that currently distributes farm inputs (e. g. fertilizer, pesticide) throughout

Honduras.

The specific objectives of the study are to: a) document the characteristics of the
existing “certified” bean seed system, including the activities of all actors (i.e., foundation
seed producer (Zamorano), “certified” seed sellers (Zamorano and Hondugenet), “certified”
seed multipliers (contracted farmers), DICTA (Directorate of Research, Science and
Agricultural Technology), and government policies related to seed production); b) assess the
farm-level demand for “Zamorano Supervised” seed; c) identify potential participants in a
new bean seed system and the conditions under which each would be willing to participate;
d) develop a prototype budget that estimates cost of producing “Zamorano Supervised” bean
seed; and e) identify government policies and actions needed to create a sustainable private-
sector based improved bean seed production and marketing system.

The increased production (multiplication) of seed of modern bean varieties that have
characteristics acceptable to farmers and the market, promoting them widely, and marketing
them to farmers in quantities and locations near their place of residence will contribute to
increasing farmer adoption of these varieties. However, the feasibility of creating the
proposed bean seed system will depend on the willingness of agricultural input firms,
research agencies, farmers, and government to support such an initiative.

This study will analyze data and information required to docmnent the
characteristics of the existing bean seed subsector (including strengths and weaknesses)
and draw on insights gained to propose a structure for a new bean seed system-- including
the possible actors, the activities that each would perform, and the conditions under

which they would be willing to participate.

1.2 Research Questions
The sustainability of a commercial seed system requires that there exist a farrn-
level demand for improved seed and the continuous seed supply of these modern
varieties. This study collected information/data to answer the following research
questions.
1.2.1 Demand for Modern Bean Varieties
1.2.1.1 Farmer
1. Varietal Chgacteristics. What are the varieties’ characteristics that farmers desire
and are these being met by the available modem varieties?
2. Willinggess to Pay. How much are farmers willing to pay for modern bean
varieties?
3. Bag Size. What quantity of seed (1'. e. bag size, lbs) do farmers prefer to purchase?
4. Seed Repurchasing. For how many years do farmers plant the same seed and how
often are they willing to repurchase fresh seed from the formal market?
1.2.1.2 Bean Traders
3. Price Discounts. What is the relative price that traders pay to farmers for
modern, compared to traditional (landrace) varieties?
4. Relative Volumes Traded. What percentage of bean sales are accounted for by

modern versus traditional varieties?

1.2.2 Supply of Modern Bean Varieties

1.2.2.1 Research Agencies
1. Price 6f Foundation Seed. What is the price at which Zamorano would be willing
to sell foundation seed to seed multipliers?
2. Necessm Resources and Conditions. What resources and conditions (e.g.
contracts) do research agencies believe are necessary for seed multipliers to produce good
quality seed?
3. Training and Supervision. How many seed farmer groups should be trained and
supervised to grow commercial seed‘, who could provide the required training, and what
should be the components of the training program?
4. Qost of Training and Supervision. How much would the institution providing
training to the seed multiplier charge for training and supervision?
5. Institutional Arrangements. What requirements and arrangements would have to be
negotiated in order for the training institution to agree to supervise the farmer seed
multipliers?

1.2.2.2 Bean Seed Growers (Previous)
1. Production Costs. In the past, how much has it cost seed multipliers to produce
improved bean seed?
2. Incentives to Prodgce Seed. What price did these farmers receive for producing

commercial seed?

 

‘Commercial seed is a category of seed where field inspections are not needed and only
purity and germination rates are checked.

3. Qthrait Have seed multipliers been given contracts to multiply seed needed and
if so, what were the terms of these contracts?
4. Necessm Resources and Conditions. What resources have seed multipliers required
to produce good quality seed?
5. 13%. What inputs have seed multipliers applied, where did they obtain them,
and what were the purchasing terms?
6. Certification. What problems have multipliers encountered in obtaining
government certification of their seed?
7. Partigipation in the New Scheme. Are previous multipliers interested in
participating in the new scheme?

1.2.2.3 Bean Seed Growers (Potential)
1. Potential Bean Seed @wers. Which farmers or group of small-scale farmers
have the necessary resources and conditions (e.g., expertise, accessible location) or with
little investment, would be capable of producing commercial seed?
2. Production Advantages. What are the advantages and disadvantages (e.g.,
production costs, credibility,) of contracting small-scale versus commercial farmers to
produce commercial seed?
3. Production Qosts. What is the estimated production cost of contracting small-
scale versus commercial farmers to multiply seed?
4. Incentives to Produce Seed. What is the minimum seed price that small-scale

farmers would require to produce commercial bean seed?

5. m. Are contracts with small-scale seed multipliers required and, if so, what
are the terms that they would require to participate in the proposed commercial bean seed
production scheme?

1.2.2.4 Government Agencies
1. y_91r1_m_e. What is the annual quantity of bean seed produced by government
agencies?
2. Contracts. What are the contract terms that government agencies offer to bean seed
multipliers?
3. Characteristics of Bean Seed Growers. What resources and conditions (e. g. contract
terms) do government agencies provide to its bean seed multipliers?
4. Seed Certification Requirements. What are the requirements that multipliers must
meet for their seed to be certified by a government agency as certified seed versus
commercial seed?
5. Agency Constraints. What problems do government seed certification agency
encounters (e.g., understaffed, resources, cars, equipment, funding) that may constrain
their ability to certify multipliers’ seed?
6. Capacity. What is the government seed certification agencies’ capacity (e.g.,
number of staff) to certify seed grown by seed multipliers?

1.2.2.5 Non-Government Organizations
1. _D_e_n_1;_agc1. What has been the magnitude of the NGOs’ demand for certified bean

seed?

2. _B1rg_$12_e. What bag size do NGOs prefer for distributing/selling bean seed to
farmers?
3. Bgngficigy Terms. What are the terms under which NGOs provide bean seed to
farmers?
4. Qharacteristics of bean seed growers. What resources have NGO artisan bean
seed multipliers had to multiply seed and which resources were provided by the NGO?
5. Contracts. What contract terms have NGOs’ offered to artisan bean seed multipliers?
6. Production Costs. What are NGOs’ estimated production cost for producing
commercial seed?

1.2.2.6 Private Sector Bean Seed Firms
1. 196mg. What is the annual quantity of certified/commercial bean seed produced by
private seed firms?
2. Bag Size. What is the current bag size, and would these firms consider marketing
beans in other size bags?

3. Productign Site. Where do these firms produce their certified/commercial bean seed?

4. I_n_p_r1t_§. Do these firms provide inputs to their seed multipliers and what are the
purchasing terms?

5. Contracts. Do these firms offer contracts to seed multipliers, and if yes, what are the
contract terms?

6. Characteristics of Bean Seed Growers. What resources (e.g., irrigation) and

conditions (e.g., provide inputs) do these firms provide to their seed multipliers?

7. 2m. At what price do these firms sell certified and commercial seed and how do
they determine the price?
8. Eroduction Costs. What are these firms production costs for multiplying seed?
9. Marketing. How do these private firms market bean seed and at what location do
they sell it?
10. Partigipation in Scheme. Would these firms be willing to participate in the
proposed bean seed multiplication scheme and under what conditions and terms would
they sell “Zamorano Supervised” seed?
1 1. Foundation Seed. From whom do these firms obtain foundation seed and do they sell
it to their seed multipliers or provide it at no cost?

1.2.2.7 Input Marketing Firms
1. LIIELIIE- What inputs do these private input marketing firms sell, and where and how
do they distribute these inputs?
2. Bean Seed. What bean varieties do input marketing firms sell, in what amounts,
and in which regions of Honduras do they sell these varieties?
3. Participatign in Scheme. Would these firms be interested in participating in the
proposed seed distribution scheme and what terms would they require to sell bean seed
through their sales network (e.g., commercial seed, bag size, volume, exclusive seller,
contract, markup)?

1.2.2.8 Bank Officials
l. Histog of Loans. Do commercial banks make loans to bean farmers, and if yes, what

is the total number of loans, value, and average value of these loans?

10

2. W. What is the minimum/maximum amount of money that commercial
banks will loan to a bean farmer?
3. Loan Terms. What terms (e.g., interest rate, collateral, amount, repayment time) do
commercial banks offer to bean farmers and would they offer special terms to farmers
participating in the proposed seed multiplication scheme?
1.3 Thesis Outline

This study is divided into seven chapters. Chapter I introduces the research
problem and presents the research objectives and research questions. Chapter 11 presents
the literature review and conceptual framework, which provide the rationale for the study
objectives, research approach, and research questions. Chapter III describes the
methodology used to collect data and methods of analysis, including the models used to
test the hypotheses. Chapter IV reports the findings on the bean seed supply in Honduras
and supporting related agents that supports the seed supply. Chapter V reports results on
bean seed demand based on a farmer survey. Chapter VI proposes a new seed system
framework and the mechanisms required to insure coordination among the agents in the
system. Chapter VII summarizes the findings, discusses policy implications, and presents

recommendations for future research.

11

Chapter II
2. Conceptual Framework

2.1 Overview

Chapter 11 presents an overview of the subsector approach, reviews the evolution
of seed systems, and seed demand, supply, and transaction costs. The subsector approach
includes the use of rapid appraisal methods to collect information and stages of seed
development. The types of seed systems include formal and informal systems. The seed
demand section discusses the structure of the random utility model that will be estimated
using a double-hurdle analysis. The seed supply section presents a budget to determine
the minimum price farmers will be willing to sell bean seed. Finally, Chapter H discusses
the nature of transaction costs and the implications that they have on farmer seed demand
and supply.
2.2 Subsector Analysis

Shaffer (1968) defines a subsector as "the vertical set of activities in the
production and distribution of a closely related set of commodities". In bean seed
production, several activities (e. g. breeding, multiplication, processing) are required to
obtain the final product, seed. Staatz (1997) recommends using subsector analysis to
analyze and evaluate how the productivity of production and distribution activities for a
good or closely related goods can be increased.
In applying the subsector research approach, it is necessary to take into

account the vertical coordination and competitive relationships in a food sector (Shaffer,

1973). All activities, from breeding to the actual sale of the seed, are governed by two

12

systems: the physical transformation system and the coordinating system (Shaffer, 1973).
In the case of bean seed, physical transformation begins when the breeder combines the
desired mix of genes, followed by multiplication, processing, and packaging of the
improved seed, and ends with transportation to the place of sale. These activities are all
controlled by the coordinating system, which include the administrative processes within
the firm and the market processes, all governed by practices and institutions involved
(Shaffer, 1973). Since there is dynamics in the way in which these systems interact, the
rules and regulations that monitor the system must evolve as the system change in order
to decrease transaction costs (Shaffer, 1973).

Data required to carry out a subsector analysis can be collected using rapid
appraisal methods. Rapid appraisal, a methodology that falls within a continuum of
informal and formal modes of data collection, leads to an objective reality (Kumar, 1993).
Rapid appraisal relies heavily on key informant interviews with individuals who have
expert knowledge about the subject of interest. During these interviews, the researcher
solicits information from the expert on a series of themes and asks the expert to elaborate
on his/her responses. Potential key informants in the food system include farmers, first
handlers, wholesale traders, processors, storage and transport agents, and distributors
(Holtzrnan, 1993).

2.3 Seed Systems
2.3.1 Stages of Seed Development
Before proposing policies to enhance the productivity of a seed system, it

is necessary to understand the current stage of seed development in a country. Seed

13

industry development can be separated into four stages that transcends from the simple
distribution of seed from farmer-to-farmer, which is regulated by customary law, to a
more complex system where private firms predominate and which is regulated by formal
institutions (Douglas, 1980; Rusike and Eicher, 1997). Douglas defines these four stages
as: l) the traditional stage, 2) the emergence stage, 3) the growth stage, and 4) the
maturity stage.

The “traditional stage” is characterized by production and exchange of seed
among farmers (Rusike and Eicher, 1997). In addition, farmer organizations (Rusike and
Eicher, 1997) or a plant breeding department (Douglas, 1980) may be involved in testing,
selecting, and multiplying small quantities of seed. However, since regulatory
organizations have not yet been established, informal habit and customary law prevail.
Thus, seed quality is the principal problem (Rusike and Eicher, 1997).

In the “emerging stage”, the government is formally involved in conducting
research and ensuring seed quality. In this stage, the government delegates seed
multiplication to selected farmers and establishes basic regulations for seed exchange
(Rusike and Eicher, 1997). However, seed of modern varieties is not widely available
(Douglas, 1980).

In the “growth stage”, as the market grows, private firms become increasingly
responsible for seed production. In response to the growth in seed demand, private firms
create new generic technologies (Rusike and Eicher, 1997) and governments establish

seed policies (Douglas, 1980). At this stage, private firms invest heavily in seed

14

production and marketing. Thus, private firms are mainly concerned about farmer
demand for seed and financial uncertainty if seed is not sold (Rusike and Eicher, 1997).

In the “maturity stage”, variety proprietorship is recognized by the government. A
main concern for private firms developing new varieties is obtaining patents (Rusike and
Eicher, 1997). Product labels become increasingly important (Rusike and Eicher, 1997)
and seed firms put greater emphasis on marketing strategies because of the
competitiveness among seed firms (Douglas, 1980).

2.3.2 Types of seed systems

Seed provision can be divided in formal and informal systems. Informal seed
systems include farmer-saved seed, farmer-to-farmer exchange, farmers buying market
grain for seed, and seed supplied by artisan seed projects. While artisan seed systems are
more organized than the former, artisan seed is not labeled as “commercial” or “certified”
seed by an authorized organization. In contrast, formal systems include commercial or
certified seed, which is supplied by commercial input distributors.

While farmers in most developing countries rely primarily on saved seed
(especially for self-pollinated crops), seed companies are playing an increasingly
important role in seed provision. For example, while farmers in East Africa rely on saved
seed as their primary source, in Kenya and Tanzania seed companies are very active
(Remington et. al. , 2002). In the case of bean seed, some seed companies in Kenya
contract seed production to farmers, and provide them with inputs on credit to grow seed
under rainfed conditions. This seed is distributed through a cooperative union,

wholesalers and through governmental selling centers, which some sell the seed on credit

15

(Kamau, 1997). Kamau’s (1997) research in Kenya identified several factors which limit
the use of certified seed, including availability, quality, price of seed, and a lack of credit.

2.3.3 Informal Seed Systems in Central America

Bean farmers in Central America rely on saved seed as their main seed source. In
recent past years, artisan seed projects have played an important role in supplying seed to
Central American farmers. Nevertheless, these projects have not been sustainable. In the
late 19808 and early 19903, these projects sought to increase the supply of modern
varieties by selecting and training farmers to produce seed (V iana, 1999). While the need
to produce seed and make it accessible to farmers has been targeted by artisan seed
programs, many improvements are needed for these projects to evolve into a sustainable
seed system that most farmers will use. For example, Guatemala’s artisan seed project
(“Cooperativa Integral Agricola Santa Gertrudis”) collapsed because of poor seed. quality
(Ordofiez, 1999). To insure that seed produced by artisan seed projects is of good quality,
Chacén et. al. (1999) recommended that the seed should at least be graded as commercial
or certified seed «depending on the quality of seed produced. Another problem with
artisan seed programs is that they are funded by projects. When the project ends, the
program struggles until a new project is initiated. For example a F INNIDA-fimded
project which started at the beginning of the 19903, and was implemented together with
the Ministry of Agriculture in Nicaragua until 1995, struggled to survive until the
National bean Program (UN ISEM) (part of the National Institute of Technology Transfer

(INTA)) was established in 1996 to replace the project (Garcia, 1999). Finally, since

16

artisan seed programs have failed to involve the private sector, they have not contributed
to the development of a sustainable seed system.

2.3.4 Informal Seed Systems in Honduras

Honduras’ informal seed system has similar characteristics as the informal
systems described above, including the widespread planting of farmer-saved seed, farmer-
to-farmer exchange, farmers buying market grain for seed, and seed supplied by artisan
seed projects.

In Honduras, the artisan seed program started in 1989 with two objectives: 1) to
accelerate the diffusion of modern varieties and 2) to train technicians and farmers in
artisan seed production (Rodriguez and Viana, 1998). The activities of the program
included motivational meetings with technicians and farmers, farmer selection, technician
selection and assignment of activities, farmer and technician training in variety selection
and crop growth, harvesting, and commercialization (marketing). The foundation seed,
which was provided by the “Prograrna Nacional de Frijol”, was produced with leader
farmers and then sold to participants in the artisan seed project. Farmers participating
sold the seed they produced to input firms and to the Directorate of Research, Science and
Agricultural Technology (DICTA), who then sold the seed to farmers on credit
(Rodriguez and Viana, 1998). According to Bemsten and Mainville (1999), while artisan
seed programs have successfully produced seed, they have failed to develop a strategy for
marketing their seed to farmers in neighboring villages.

In the past few years, NGOs and development projects have distributed certified

seed to farmers, especially after natural disasters. NGOs and projects give the seed to

17

farmers for free or ask farmers to repay in kind after the harvest. Consequently, after
natural disasters input dealers accumulate a large stock of seed which they cannot sell. In
addition, seed firms reported that even in normal years, NGOs free seed distribution
initiatives reduced commercial seed sales.

Recognizing the need to avoid undermining the development of a private seed
sector, several projects in Afiica have provided seed vouchers (i. e. coupons or certificates
issued by the NGOs or projects) to farmers that guarantees a cash value for the exchange
of seed from approved sellers (Remington er. a1. , 2002). In addition, Remington eta].
(2002), describes the success of seed fairs in several countries in Africa; where
commercial seed companies, input dealers, and market grain traders offer their seed to
farmers under the supervision of a fair committee that recommends a sale price and buys
the vouchers back from the sellers. While provision of free seed to farmers undermines
the development of a commercial seed system, the voucher approach actively involves the
seed supplying agents and helps to increase the likelihood that farmers will seek to
purchase seed from those agents in the future.

2.4 Function of Seed

Tripp (1997) states that seed has two functions. First, as an input and second, as a
source of germ plasm. Tripp emphasizes that farmer demand for seed will differ,
depending on which need he/she wants to satisfy (Tripp, 1997).

Farmers need seed as an input when natural disasters occur (e.g. hurricanes) and
the farmer is left without seed to plant. In this case, the farmer will purchase any

available seed, without regard to his/her variety preferences. In addition, due to poverty,

18

farmers in developing countries are often forced to obtain seed from outside their farm
because they are unable to harvest and save sufficient seed for the following season
(Tripp, 1997). Alternatively, a farmer who wants a specific variety might purchase a
small quantity and multiply it himself to obtain seed with the characteristics that he/she
desires. However, this is only possible for self-pollinating crops (because of the slow rate
of genetic degradation) and crops multiplied via asexual reproduction.

On the other hand, there are two reasons why a farmer may demand germplasm.
First, farmers may wish to obtain seed of a new variety that is higher yielding and or more
pest/disease resistant than the farmer’s current variety. Second, farmers may wish to
obtain gerrnplasm of the same variety to replace retained seed because its production
potential has deteriorated due to a breakdown of its pest or disease resistance (Heisey and
Brennan, 1991; Tripp, 1997).

In order to develop new varieties that meet farrners’preferences, researchers
should observe the varieties that farmers plant to determine if they have been adopted
because of farmers’ preferences or because farmers desire seed just as an input.
However, farmer demand for modern varieties also depends on the performance of the
varieties in farmers’ fields (relative to traditional varieties (TVs)), the acceptability of
these varieties in the market, their local availability, and their affordability (Bemsten and
Mainville, 1999).
2.5 Seed Demand

Farmer demand for seed depends on several factors including seed rate (planting

density), area planted, and varietal choice.

19

2.5.1 Seed Rate

In a competitive market for a single factor the neoclassic input demand (e. g. seed
rate) function is the inverse of the marginal value productivity function, from the input
level of maximum average value product to higher levels (Beatie and Taylor, 1993)
(Figure 1). The optimal seed rate per hectare is found at the point where the marginal
change in yield, due to an increase in planting rate, is equal to the seed-to-grain price
ratio. The optimal seed rate is determined by testing several planting densities by
varying the distances between plants and rows. However, farmers in Honduras usually
plant the same amount of seed per unit area regardless of variety, as long as there is no
big difference in bean grain size2- although farmers in Honduras use a lower seed rate in
the Primera than in the Postrera, due to higher rainfall in the Primera.

Figure]. Factor demand function for single-variable factor.

AVP

 

 

* MVP = marginal value product
AVP = average value product

 

2 Farmers in Honduras usually use the same seed rate for bush-like plants. This could be
attributed to custom or feeling secure of the planting practice because of past experience.

20

2.5.2 Area Planted

Farmers in Honduras usually plant the same hectarage to beans every year. This is
because of limitations in land and access to credit. Thus, the total quantity of seed a
farmer plants does not vary greatly from year-to-year.

2.5.3 Varietal Choice

Farmers often plant different varieties from year—to-year. In such instances,
farmers face the decision of choosing between two inputs (varieties) with different prices
that yield two products with different attributes and prices. The farmers’ decision
regarding which variety to plant can be reduced to a breakeven price analysis. In
business management analysis, the breakeven point is the point where total costs are
equal to total revenues (Gardner, 1948). In the context of comparing two varieties, the
breakeven price of seed is the maximum seed price that makes the profits of using either
variety equal. In other words, the breakeven price is the highest price a farmer would pay
for seed of a variety, given the output prices, seed price of the compared variety, and
yields of both varieties. Thus, a farmer would only be willing to buy a new variety if the
net revenue from that variety is equal to or greater than his/her existing variety.

2.5.4 Hedonic Price Model

Hedonic pricing can be used to estimate the price that farmers would be willing to
pay for a product that is yet to be introduced in the market. The hedonic pricing method
is based on the idea that price is a function of a product’s attributes (Jimenez and Oppen,
1999). In the case of seed, farmers will be willing to pay a high price for a variety if

he/she likes the cryptic and noncryptic characteristics that it possess. Cryptic

21

characteristics are the inner properties of the variety, which are invisible (e. g. nutritional
and productive properties). In contrast, a varieties non-cryptic characteristics are
properties that can be assessed by touch and sight (e. g. weight, color and shape). A model

representing this idea is shown in Eq. 1
rmv= f( c, n) (Eq 1)

where 1‘ v = price of the modern variety

m

c = cryptic characteristics

It non-cryptic characteristics

Hedonic pricing does not contradict the concept that price is a function of supply
and demand. However, it does assume a competitive equilibrium in a plane of several
dimensions, where buyers and sellers are located (Jimenez and Oppen, 1999). In this
equilibrium, farmers’ characteristics like capital (e.g. land), human capital, labor, and
infrastructure are held constant, which in reality may not be true. In an attempt to
explain that variability, variables that measure farmer (i.e. seed buyer) characteristics can
be included.

While breakeven analysis only takes into account yields, seed prices, and
grain prices, hedonic pricing analysis also incorporates varietal seed characteristics as
explanatory variables. If the model is used to compare only two varieties, these varieties
can be thought to be a modern and a currently used variety. Using hedonic price analysis,
the highest price a farmer will pay for a variety can be expressed as a function of the

attributes of the variety itself, and the attributes of other varieties, as follows:

22

rmv = f(CCV ’ cmv’ nCV ’ n'l‘l'lV) Eq' 2

where rmv = price of the modern variety

cCV = cryptic characteristics of current variety

cmv = cryptic characteristics of modern variety

ncv = non-cryptic characteristics of current variety

nmv = non-cryptic characteristics of modern variety

In this hedonistic price model, cryptic characteristics include yield and other
nonobservable characteristics of the modern and the farmers’ current variety. Farmers
assessment of the non-cryptic attributes (e. g. size, color, shape) are reflected in farmers’
expectations regarding the price that they think the variety would sell for in the market.
In other words, rather than incorporating output prices as explanatory variables, price is
reflected in the farmers’ valuation of each characteristic. Furthermore, if factors that are
held constant in the Eq. 2 ( e. g. capital, human capital and labor—since they do not appear
in the model) are introduced into the model, the price of a modern variety is a function of
the attributes of the modem variety, attributes of the farmers’ variety and socio-economic

characteristics of the farmers. Thus, the new model may be represented as:

va = f(ccv , cmv, ncv , nmv, fr) Eq. 3
where “fr” takes into account the farmer’s management and socioeconomic characteristics

and the other variables are the same as shown in Eq 2.

23

In summary, the price of bean seed is a function of its yield, farmers’ varietal
preferences (e.g. seed size, shape, color, cooking time, and palatable characteristics of
the grain), and farmers’ socioeconomic characteristics3 (Figure 2).

Limited—dependent variable models are typically used to determine the influence
of factors (variables) that explain purchasing decisions (i.e. willingness to pay). These
models may include truncated variables (i.e. values above or below some threshold level
are not observed due to either research design or naturally). According to Roosen
et.al.(1998), a double hurdle model is appropriate for evaluating data with truncated
dependent variables such as willingness to pay (WTP) data sets.

Figure 2. Bean price factor dependence.

Price
Varieties' production Consumers' preferences

characteristics (1. e. yield)

00115111110“. . Varietal Characteristics
socroeconotmc (re. seed size. shape)
characteristics

3 The farmer’s characteristics include education (Heisey and Brennan, 1991),
management practices they use, capital and age.

24

A double-hurdle model is estimated in two stages. First, bivariate probit analysis
is used to estimate the first hurdle and then truncated regression analysis is used to
estimate the second hurdle (Cragg, 1971; Back, 2004).

2.5.5 Willingness to Pay

Myrick (1993) defines willingness to pay as the maximum amount of money an
individual is willing to pay, instead of doing without the increase in the quantity of some
good. While willingness to pay is most frequently associated with nonmarket valuation
techniques like contingent valuation (Hanley et. al., 1998), it can also be used with
observed data of marketed goods, where WTP is a limited dependent variable (Roosen
et. a1. ,1998). The demand curve provides information on the marginal willingness to pay
(Perman et. al., 1999).

One way4 to structure the deterministic part of a random utility model (e. g.
double hurdle model) is by specifying the utility of an individual as a function of a
combination of that individual’s characteristics and attributes of an alternative good (i.e.
improved bean seed) (Kennedy, 1998). Thus, a random utility model can be used to
estimate the farmers’ highest price that a farmer would pay for seed of modern varieties,
given certain information about the traits of the modern varieties, farmers information
about traditional varieties (i.e. yield, cooking characteristics, and grain color, size and
shape), and information about the farmer per se (i.e. capital, labor). In a random utility

model, the farmer’s latent value ( i.e. breakeven price or highest price that a farmer

 

4 The two other ways to structure the deterministic part of the random utility model is by
1) specifying the utility of the individual as a linear function of that individual’s
characteristics or 2)as a linear function of the attributes of the alternative.

25

would pay for seed of modern varieties) is not estimated and average yields of farmer’s
landrace varieties are not observed as they are in a break even price analysis. Instead, we
collect data of farmers’ affrrrnative or negative response to the purchase decision and
WTP. By specifying the latent value (i.e. breakeven price) as a function of attributes of
the product and characteristics of a farmer, it is possible to observe how an attribute or a
characteristic increases or decreases the probability to purchase and WTP.

A farmer’s buying decision has two dimensions--the decision to purchase
and willingness to pay a given price. An advantage of the double hurdle model, as
opposed to the Tobit model, is that it allows for the possibility of separating these two
decisions (Burton et. al. , 1994). The double-hurdle model assumes that the farmer
makes these two decisions separately—a farmer first decides to purchase a new variety to
replace retained seed of an old variety and later decides on the maximum price that he/she
would be willing to pay, depending on his assessment of the benefits of planting the
improved seed. The farmer replaces his/her retained seed because he/she is convinced
that modern varieties have superior characteristics over the retained seed, or just to try a
new variety to observe how it responds.

2.6 Seed Supply
Seed supply information is important because it enables to understand the price
determination and to assess the possibility of new firms entering the bean seed subsector.
2.6.1 Supply function
The neoclassic supply function specifies a schedule of prices and quantities of

product that a firm will supply when investments have already been made. In a

26

competitive product and input market, the supply function is the inverse of the marginal
cost function from the minimum average variable cost point to higher values (Beatie and
Taylor, 1993)(Figure 3). In the short run, the minimum price that the firm will be willing
to sell its product is at the point where rrrinirnurn average variable cost is minimized
(Point A in figure 3). In the long run, a firm will be willing to sell its products at the
point where minimum average total cost is minimized (Point B in figure 3). However, to
decide on the minimum price that a firm is willing to sell the product before the required
investments have been made, budgeting techniques can be used.

Figure 3. Short-run firms’ supply curve in a competitive market.

3 MC
ATC
AVC
B
A

 

 

* MC = Marginal cost
ATC = Average total cost
AVC =Average variable cost
2.6.2 Budgeting
Three major types of budgets are used to analyze farm operations. These are the

whole farm budget, enterprise budget, and partial budget. The enterprise budget is used

27

when we need to analyze the profitability of a production system or activity. The
enterprise budget consists of a list of revenues and expenses incurred in the production of
a product (Doye, 2005). In particular, for bean seed production, an enterprise budget
consists of list of expenses like land preparation, inputs, labor expenses for applying
inputs, and processing of the seed. The enterprise budget will provide the cost of seed.
2.7 Transaction Costs

2.7.1 The Nature of Transaction Costs

North (1990) defines transaction costs as the costs involved in an exchange,
including the costs of information, protecting rights, policing, and enforcing agreements.
On the demand side (e.g. seed buyer), Rusike and Eicher (1997) stated that “the
transaction costs are caused by imperfect information, as well as transportation,
negotiation, motivation, monitoring and supervision, and contract enforcement”. On the
supply side (e.g. seed grower), transaction costs include the costs of gathering demand
information (North,1990), asset specificity’ (Williamson, 1985), principal agency
problems‘5 (HayamiY., and Otsuka K., 1993) , and uncertainty in the market (Nicholson,
1998). With respect to seed, the incremental revenues expected from a modern variety,
either by supplying it (firm) or adopting it (farmer), must exceed the incremental costs of

supplying or obtaining the modern variety, including transaction costs.

 

’ Difference between assets value in its optimal use and its second best alternative.

6 Problems in contracting due to information asymmetry like moral hazard and adverse
selection

28

2.7.2 .Farmers’ Transaction Costs

On the demand side, farmers’ information costs include the costs searching for the
characteristics that he/shc values in a bean variety. Seed characteristics include
noncryptic (visible) characteristics and performance characteristics of the variety. To
assess performance, farmers need to observe the variety, either in their field or in a
demonstration plot of a promoting organization, the former being most trusted by
farmers. Because it is difficult for farmers to measure all of these characteristics for
several varieties at a time, government agencies and NGOs establish demonstration plots
in farmers fields for farmers to assess these attributes. Nevertheless, the final test of a
variety is when a farmer plants it in their own field using his/her technology and
consumes the grain to measure its edible characteristics. Finally, transportation costs
from the farrngate to the seed retail store play a role in the farmer’s decision to buy seed,
specially for farmers living in remote areas.

The transaction costs associated with protecting rights (e. g. writing the contract),
policing (e.g. monitoring), and enforcing (e.g. legal fees) agreements also affect a
farmer’s decision to buy seed. In the first stages of seed system development, farmers
save their own seed or obtain it seed from another farmer. Thus, seed inspection is easy
and the trust among farmers is high. However, when farmers buy seed from a
commercial seed distributor, they are not able to inspect the seed. Rather, they must rely
on the govemment’s seed certification unit to guarantee seed quality (Rusike and Eicher,
1997). F urtherrnore, if the seed fails to meet the required quality characteristics (e.g.

purity, germination rate), the farmers’ power to enforce an agreement ( i. e., obtain

29

compensation) is low because they usually do not have the means to hire a lawyer to
protect his/her rights. And even if a farmer made a claim to the seed retailer, the retailer
would only replace the purchased seed; which would not fully compensate the farmer,
since he/she would have lost other inputs previously applied and possibly would have
missed the proper time to plant his/her crop. Consequently, to minimize these transaction
costs the government must establish a reliable seed certification system, as well as the
other institutions required to protect rights, police and enforce agreements in order to
insure seed quality.

Thus, for a farmer to adopt a modern bean variety, the benefits (e.g. higher yields,
higher prices) must be greater than the incremental cost due to the higher seed price and
the transaction costs aforementioned.

2.7.3 Seed Firms’ Transaction Costs

On the supply side, firms face three problems: gathering demand information,
recovering costs, principal-agent problems, and uncertainty in the market. The cost of
gathering demand information could be very costly for a small seed firm especially in the
first stages of seed development. This information is important to make decisions on the
amount of capital to set for research and development. According to Rusike and Eicher
(1997), in the first deve10pment stages, private finns-- which must invest in asset-specific
equipment, research and promotion-- usually do not recover their investments. To reduce
these risks, a firm might, for example, rely on the government to conduct varietal research

and only multiply and seed these public varieties.

3O

The principal-agency problems could also arise in a bean supply scheme in which
the contractor (principal) or the entity in charge of selling a category of seed (e.g.
foundation, certified, commercial) contracts a farmer (agent) for seed multiplication and
the agent tries to minimize its costs by using fewer inputs, thus, producing lower quality
seed. This problem is complicated by the fact that sometimes, it is not possible to assess
seed quality until after the crop is planted (i.e. seed-home diseases).

Uncertainty in the market (e. g. bean prices, demand for beans) leads to inefficient
allocation of resources (Nicholson, 1998). That is because a bean multiplier decides on
courses of action (e.g. input use, investment) based on the probability of events to occur
and not on the precise event. This leads to lower profits and even bankruptcy, as depicted
in the example of Rusike and Eicher (1997).

2.7.4 Government’s Role in Reducing Transaction Costs

Because transaction costs plays an integral role in the farmers’ decision to adopt a
variety and a firms decision to supply a variety, the government or a foreign organization
often play a facilitating role to help support the evolution of the seed systems stages
(Douglas, 1980; Rusike and Eicher, 1997). According to Fafchamps, Janvry and
Sadoulet (1995), the government needs to establish policies to promote technological
change, reduce transaction costs, lower credit costs, and relax credit constraints that
farmers face in order to enhance competitiveness and thereby seed system development.

2.8 Summary

The subsector approach gives a useful frame to explain the vertical set of

activities in the production and distribution of closely related set of commodities. Rapid

31

appraisal methods are used to collect data to carry out subsector analysis. The diagnostic
of the subsector analysis on the bean subsector should identify the stage of seed
development for later proposing of policies to enhance the productivity of the bean
system.

Developing countries in Africa and Latin America have used the informal seed
systems as the primary source of obtaining seed. This informal seed system gives a poor
quality seed since the seed is coming from farmer-to-farmer and farmer-saved seed, or are
of good quality but not sustainable like the artisan seed projects have been. The failure of
artisan seed projects has been the inability to develop a strategy for marketing their seed.

The breakeven analysis is used in farmers’ decision on which variety to plant
when the demand for seed is static, as when the farmer plants same area and does not vary
the quantity of seed per area. Instead of including the output price of a product, which
encapsulates the characteristics of a product, as a explanatory variable, hedonic pricing
includes the characteristics of the product as explanatory variables to capture the value of
each characteristic. Factors that are held constant in hedonic pricing like capital, labor,
and infrastructure, and consumer (farmers) characteristics can be included in a random
utility model to explain purchase decision and willingness to pay. A random utility
model is analyzed with a double-hurdle approach. The first hurdle with a probit model to
analyze the purchase decision and the second with a truncated regression to analyze
willingness to pay.

A bean seed budget is used to determine the minimum price farmers will be

willing to sell bean seed.

32

Transaction costs or costs involved in an exchange include costs of information,
protecting rights, policing, and enforcing agreements. Information is needed by farmers
of the varieties they are offered, and by firms of farmers preferences. Protection of rights,
policing and enforcing agreements are an important part of the decision of the farmer
since if there is no system that could ensure these rights, the transaction costs could deter
farmers’ purchase decision. On the supply side, these rights are also important when
dealing with contractees where principal-agency problems arise, with competitors when

intellectual property rights are not in place and with customers.

33

Chapter III.
3. Research Methods

A rapid appraisal was conducted to collect information about the bean seed
subsector in Honduras. During the rapid appraisal, key informant interviews were
conducted and secondary data were collected. In addition, a sample of bean farmers were
surveyed to collect data required to assess farmer demand for modern bean varieties.
Also, a budgeting model was used to estimate the farm level profitability of producing
commercial seed. Finally, a willingness to pay analysis, using a double-hurdle model,
was used to assess farmer demand for modern varieties.
3.1 Data Collection

3.1.1 Rapid Appraisal

The rapid appraisal method was used to collect data needed to characterize the
bean seed subsector. Key informant interviews were conducted with staff of input
dealers, seed firms, NGOs and projects, banks, research agencies, government offices;
and farmer groups and seed growers.

3.1.1.1 Input Dealers

Many firms in Honduras market agricultural inputs (e. g. seed, fertilizer,
insecticide). However, only three dealers were contacted—DUWEST, FERTICA and
CADELGA-- because these firms own or do business with many retail stores around the
country. Thus, they could possibly be utilized in the future to market bean seed. Initial
attempts to contact CADELGA and FERTICA revealed that they were not interested in

participating in the bean seed scheme. However, staff of DUWEST agreed to be

34

interviewed. During the meeting, information was obtained regarding the type of inputs
they sold and where they distributed inputs. After explaining the proposed bean scheme,
DUWEST staff were asked under what conditions they would be willing to participate
and sell bean seed through their sales network.
3.1.1.2 Seed Firms

Only two private firms produce bean seed in Honduras, Zamorano and
Hondugenet. Zamorano is a private agricultural college. Hondugenet is a private
company, which was a parastatal before it was privatized in the early 1990s under the
Law of Agricultural Modernization. Staff of Zamorano and Hondugenet agreed to be
interviewed and provide information about their bean seed production strategies,
varieties produced, problems encountered, bean seed production contracts, seed
production locations, inputs used, and marketing strategies.

3.1.1.3 NGOs and Projects

In recent years, several NGOs and projects have been involved in the bean seed
production and distribution, especially following Hurricane Mitch (1998). In 2000,
USAID funded a major project (Post Mitch Revitalization of the Agricultural Sector) to
produce and distribute bean seed. Staff of two NGOs «World Neighbors (WN) and the
Catholic Relief Services (CRS)— and two projects «the Development Project of the
Mideast (PRODERCO) and the International Center of Tropical Agrigulture’s (CIAT)
seed project« were interviewed because of their extensive participation in seed
distribution schemes. In 2000, CIAT, in collaboration with Food and Agricultural

Organization (F A0) and Movirnento Liberazione e Sviluppo (MOVIMONDO) organized

35

the “Production and Commercialization of Improved Artisan Seed” project, which
worked with seven NGOs, including World Neighbors. Under the Post-Mitch project,
World Neighbors, CR8, and PRODERCO worked in collaboration with Zamorano to
distribute seed to small farmers in the Departments of El Paraiso, Francisco Morazan, and
Olancho. These NGOs were contacted to collect information about their bean seed
production and commercialization/distribution activities, the demand for improved seed,
their distribution strategies, and their interest in participating in the proposed bean seed
production scheme.
3.1.1.4 Banks

Key informant interviews were conducted at four banks: Banco de Occidente,
BGA, BAMER, and the National Agricultural Development Bank (BANADESA). Banco
de Occidente, BGA, and BAMER were contacted because they are major participants in
the capital market and have branches throughout Honduras. BAN ADESA was contacted
because it was established by the government to support farmers. Bank staff were
interviewed to collect information about their loan terms, volume of past loans to bean
producers, loan services available to bean farmers, their interest in lending to bean seed
producers, and the conditions under which they would extend loans to bean seed farmers.

3.1.1.5 Research Agencies

Two research agencies were contacted: the Bean Research Program (PIF) at the
Panamerican Agricultural School and The Directorate of Research in Science and
Agricultural Technology (DICTA), which are both responsible for developing modern

bean varieties in Honduras. Staff from these agencies were interviewed to collect

36

information about their bean seed production capacity, varieties released, prices of
modern varieties, resources and conditions that farmers must follow to produce high
quality bean seed, the structure of the existing bean seed production system, and the
ability of these agencies to train farmers in bean seed production.
3.1.1.6 Seed Certification Agency

CERTISEM, the agency in charge of certifying seed, dictates and enforces grain
quality standards for seed marketing. CERTISEM staff were interviewed to collect
information about the quality requirements for seed certification, the methodology used to
measure these requirements, the legal requirements that must be followed to initiate the
proposed bean seed scheme, and to determine the requirements for marketing bean seed.

3.1.1.7 Farmer Groups

Two groups of farmers and an individual bean seed contractee were interviewed:
farmers in Yaruca, Atlantida; farmers and managers of the Regional Association of
Agrarian Services in the Orient (ARSAGRO); and a Zamorano bean seed contractee. The
Yaruca farmers were interviewed because they have a potential to grow bean seed in the
Apante rainy season (December to February) and have access to technical support fi'om
the Atlantic Littoral University Center. The ARSAGRO farmers and managers were
interviewed because they had previously participated in a CIAT-managed artisan bean
seed production scheme. The bean seed contractee was interviewed because of his
experience in bean seed production and knowledge of the typical problems encountered in
growing bean seed. During these interviews, data were collected about their costs of seed

production, incentives and constraints to producing seed, and the conditions under which

37

they would be interested in participating in the proposed bean seed production scheme.
The two farmer groups were interviewed to assess their willingness to participate in the
proposed bean seed scheme, to learn about their bean and bean seed production practices,
and to collect data for a budget analysis.
3.1.1.8 Traders
Market traders in “Las Americas” market in Tegucigalpa were interviewed to
obtain information on the prices they would pay farmers for several modern varieties and
assess the price discounts that they impose on those varieties.
3.1.1.9 Budget Data Collection
Data required to construct a seed production budget were collected from several
key informants. Key informants interviews were conducted with three small-scale bean
seed producers in El Paraiso Department, who are members of the farmer association
ARSAGRO’, which jointly with CIAT, produces and sells bean seed of modern varieties
to farmers in the region. Data collected from the three farmers included costs of land
preparation, inputs, labor, and seed processing. In addition, bean seed cost of production
data were obtained from World Neighbors, an NGO working in collaboration with CIAT
in the aforementioned seed production initiative. These data were extracted from the
records of ten participating farmers, including the cost of inputs, labor, seed processing,
and tool investments. Finally, training investment costs were collected through

interviews with experts (i.e. Professor Juan Carlos Rosas (Zamorano) and Rodolfo

 

7 ARSAGRO was participating in a CIAT seed scheme with the NGO ICADE.

38

Pacheco (CIAT)) in bean seed production; and data on the cost of land registration and
legal fees were collected from the government seed certification unit.

3.1.2 Farmer Survey

A farmers survey was conducted to assess the demand for modern bean varieties.
A total of 72 farmers were surveyed in El Paraiso and Olancho, the two departments in
Honduras ( Figure 4.) which have the largest bean hectarage (i.e. El Paraiso 13% and
Olancho 15%, of the total area planted in Honduras (Agricultural census, 1993)). In El
Paraiso (Figure 5.), the 24 most important bean-producing villages8 in the Department
were listed and divided into two regions: the plateau and the valley region. Each region
included twelve villages. The twelve plateau villages were Arauli, Pescadero, Linaca,
San Matias, San Geronimo, Jacaleapa, El Barro, El Tablon, Las Anirnas, Teupasenti, El
Arena], and Moroceli. The twelve valley villages were Jutiapa, Coyolar, El Obraje,
Apali, Chirinas, Sartenejas, Chichicaste, Poteca, Montailita, Quebrada Larga, Zamorano,
Zapotillo, and Matazano.

In Olancho aiigure 6), the 13 most important bean-producing villages in the
Department were listed and divided in two regions«the North West mountain region with
a total of six villages and North Central and Central valley region with 7 villages. The six
mountain villages were Yocon, La Union, Salama, Silca, Talgua, and El Rosario. The
seven valley villages were Manto, Guarizarna, San Francisco de la Paz, J irnazque,

Jutiquile, Catacarnas, and Boqueron.

 

8 Most important bean producing villages” refers to villages where small-scale farmers
are known to grow beans, as opposed to villages where a few big farmers produce most of
the beans.

39

 

 

      
 

. § ;

ffoas‘so‘ 1106' iso
III—=—

 

 

1” ‘. r" _ ‘~ 2 ‘
9 -~ ;: . g
- . - ...

 

 

 

4o

Figure 5. Department of El Paraiso, Honduras.

EI Ta DION. PotecaMOfltaDIta

Chirinos.
San JeronimooLinacaEl Coyolar

NICARAGUA

 

 

 

Figure 6. Department 3f Olancho, Honduras.

 

W "’

’El Rosario .Manto .
.Salama San Francisco de La Paz

Srlea °El Boqueron

 

 

 

 

41

Finally a sample of 13 villages, from the total of 37 villages in both Departments,
were selected; three in the plateau region of El Paraiso, four in the valley region of El
Paraiso, three in the North West mountain region of Olancho, and three in the North
Central and Central Valley Region of Olancho. Six farmers were selected in each village,
except for two villages in El Paraiso because of the proximity of two villages (i.e., Poteca
and Montafiita; 3 farmers per village). In each village (except as noted above), the
sample of six bean farmers was selected at random from lists that included 19 to 58
farmers«developed in situ with the help of two to three farmers in each village.
Information about the selected departments, villages, and farmers is summarized in

Table] .

Table 1. Number of farmers surveyed by department, region, and village in Honduras,
Bean Survey, 2002.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Department Topography/Region Village Name Village Populationa
El Paraiso Plateau Linaca 39

Plateau San J eronimo 22

Plateau El Tablon 27

Valley El Coyolar 26

Valley Clririnas 36

Valley Potecab 19

Valley Montafiita 26
Olancho North West Salama 52

North West Silca 41

North West El Rosario 58

North Central Manto 42

Central San Francisco de la Paz 57

Central El Boqueron 44
a Estimated village population developed from a farmer aided in situ village farmer list.

Six farmers selected in each village except for Poteca and Montafiita, where three

farmers were selected in each.

 

42

 

3.1.2.1 Questionnaire

The questionnaire focused on soliciting information from farmers about their
preferences for and their willingness to pay for improved bean seed. F arrners’ varietal
preferences were first identified by asking them which bean varieties they had planted and
sold during the past two years. Additional questions designed to measure farmers’
preferences and demand focused on determining their awareness and assessment of
widely available modern varieties, their assessment of new modern bean varieties (i.e.
farmers were shown samples of six recently or soon to be released varieties), and their
demand (i.e. willingness to pay) for improved bean seed.

Farmers’ awareness of widely-available modern varieties was measured by asking
them if they had planted or heard of the most recently released bean varieties, Dorado and
Tio Canela. Farmers’ assessment of these modern varieties was solicited through an
open-ended question that asked farmers to identify the good and bad qualifications of Tio
Canela and Dorado. Farmers’ assessment of the traits (i.e. grain size, shape, weight, and
color) of new modern varieties«which included already released and soon to be released
varieties«was evaluated using a likert scale (1 = excellent, 5 = poor). F arrners’ demand
for each new variety was measured by their willingness to pay. To measure their ratings
and willingness to pay, the farmers were shown six modern varieties in resealable
transparent plastic bags. Each variety was labeled with a number instead of its name, so
farmers would not respond based on his or her prior knowledge about the specific variety.

Farmers were allowed to take the bean seeds out of the bags so they could examine them

better. The six bean varieties shown to farmers were: Catrachita, Dorado, Tio Canela 75,

43

Milenio, Amadeus 77 and Carrizalito«labeled from one to six, respectively. The latter
three have not yet been released to the market. In addition, farmers were asked to
estimate the price that they thought traders would pay for each variety. Finally, farmers
were asked about their recent bean production history (e.g., area planted, yields) and
socioeconomic characteristics (e.g., age, education).

The draft questionnaire was pretested with farmers in the Zamorano Valley,
revised to clarify ambiguous questions, and finally implemented in the aforementioned
regions.

3.2 Empirical Methods
3.2.1 Empirical Model for Demand

A double-hurdle model was used to determine farmers’ interest in purchasing
MVs and their willingness to pay for these MVs. The double-hurdle approach, as the
name indicates, divides the analysis into two parts: first a probit regression and then a
truncated regression. The probit model is used to determine the degree and direction of
influence that farmers’ personal characteristics and varietal traits have on his/her purchase
decision (i.e. marginal probability of each characteristic); and the truncated regression is
used to determine the degree and direction of influence that farmers’ personal
characteristics and varietal traits have on how much a farmer is willing to spend on each
variety (i.e. marginal willingness to pay).

The following double hurdle model was used to assess farmers’ purchasing

decisions and WTP:

44

Y1 = Bo + Xik Bk + vr (Probit) for i= 1,..,72, k=1,...,10
where Vi is the error term.

WTPi = Bo + Xik 7k + ei (Truncated) for i= 1,..,72, k=1,...,10
where eit is the error term.

Yi = desirability to purchase modern variety at a price between L 9 and L 17
(1=yes, 0=otherwise)

WTP = the highest price (between L9 and L17) willing to spend on a purchase.
(willingness to pay)

Xik= set of explanatory variables determining the desirability to buy

Bk and 7k are the parameter vectors to be estimated.

Separately from the double-hurdle model, an enterprise budget was used to
estimate the cost of bean seed and to analyze the price at which bean seed can be sold.
3.3 Empirical Model for Seed Supply

A budget was used to estimate the minimum price of seed at which bean farmers
would be willing to produce bean seed. The general objective of using this model was to
determine the feasibility of producing bean seed of modern varieties at a price lower than
L 30.8 /kg (i.e., the current price of certified bean seed). Sensibility analysis was

conducted on the budget to estimate the risk borne by the farmer for producing bean seed.

45

3.4 Summary

The rapid appraisal methodology was used to collect information from key
informants required to describe the bean seed subsector in Honduras. In addition, a
sample of bean farmers were surveyed to collect data about their bean variety preferences,
including their assessment of six modern varieties and their willingness to pay for these
varieties. To assess farmer demand for modern bean varieties, a double hurdle model was
used to determine the degree and direction of influence that farmers’ and varietal
characteristics have on farmers’ purchase decision and willingness to pay. To assess the
feasibility of supplying improved bean seed, a budgeting model was used to estimate the
farm level profitability of producing commercial seed and the minimum price that the

farmers need to have an incentive to produce bean seed.

46

Chapter IV.
4. Bean Seed Supply in Honduras

This chapter describes the activities required to develop, multiply, and distribute
seed of modern bean varieties and the agencies responsible for carrying out these
activities. It also describes the participants in the Honduran seed subsector (i.e.
governmental agencies, Zamorano university, NGOs and projects, private seed firms,
input distributors, and banks).
4.1 Seed Production Stages

New seed varieties are developed through a sequence of stages (Figure 7 ). First,
plant breeders produce breeders’ seed by crossing various lines, multiply the breeder seed
to produce foundation seed, and multiply the foundation seed to produce registered seed.
Seed for sale to farmers is then produced by multiplying registered seed to produce
certified seed. Finally, certified seed may be multiplied to produce commercial seed. As
discussed below, each of these seed classes must meet specific quality standards.

Figure 7. Bean seed development stages.

Breeder seed
Foundation seed
Registered seed
Certified seed

1

Commercial seed

47

4.2 Seed Supply Agents

4.2.1 Seed Certification Service

In Honduras, certified and commercial seed production’ and distribution is

regulated by a government entity, CERTISEM (Seed Certification), which is under the

National Service of Farming Health of the Ministry of Agriculture in Honduras. Prior to
2002, a committee was in charge of creating and revising policies regarding the
production, distribution, and importation of seed. However, the committee was dissolved
in summer 2002.

Seed producers are required to follow a set of rules to be able to sell their product
as commercial or certified seed. Table 2 summarizes the steps that seed growers must
follow to meet the seed certification requirements and the cost of meeting each
requirement.

Who bears these costs depends on the arrangement between the seed producers
and the sellers. While a vertically integrated firm bears all the costs associated with the
seed certification process, a processor10 (who only sells certified seed) may negotiate a

contract with a producer under which the processor pays all of those costs.

 

9 Foundation and registered seed are mainly produced by licensed producers (DICTA,
PIF), which are not required to have their seed inspected.

'° Processors include firms or individuals that cleans and sorts seed using a machine.

48

Table 2. Steps seed industry producers and processors must follow to obtain seed

certification from CERTISEM, Honduras, 2002.

 

 

Step Cost Characteristics
1. Register as:
Individual merchant L 500 Every 2 years
Producer a) L 1,000 b) L 500 a) First time b) Every 2 years
Processor L 1,000 Every 2 years
Importer L 1,500 Every 2 years

 

2. Plot registration

L 50/mza (L 71/ha)

Each cycle/cropping season

 

3. Plot supervisionb

L2/km

Outside urban area

 

 

 

 

 

Source: CERTICEM

 

4. Seed sampling L 2/km - Outside urban areas
L 0.50/sampled bag - 45 kg bag
5. Seed label Equal to the sale i. e. If seed sells for L 1,500/qq (100
price of 1 kg of seed lbs), the label will cost L 33/qq.
a
lmz = 0.7ha

Only required for certified seed, but not for commercial seed.
Exchange rate (2002): US $ 1.00 = L 16.44

 

Seed merchants (individuals or businesses) must purchase their supplies from a

certified or a commercial seed producerl '. With respect to seed certification regulations,

a major difference between these two types of seed is that while certified seed must be

inspected both in the field and in a laboratory, commercial seed is only inspected in a

laboratory after the producer gives CERTICEM a sample of the harvested seed. In

addition, certified seed must meet higher quality standards, as specified in Table 3.

CERTISEM has established specific field specifications for growing certified seed: 1) no

more than two varieties may be grown per farm and both varieties must have

 

” While individuals or business may import seed, bean seed has only been imported once
in the past seven years. In 1998, seed was imported from Nicaragua following Hunicane
Mitch, which reduced the seed supply.

49

 

characteristics that are easy to differentiate by sight; 2) fields are not eligible in which

different bean varieties or soybeans have been grown in the prior six months; 3) a field

where bean varieties have been planted in the prior six months must be planted to the

same variety with the same or lower category of seed (e.g. cannot grow certified seed if

commercial seed was previously grown in the field); 4) all plants infected with bacterial

or fungal diseases must be removed; and 5) fields planted to different bean varieties must

be separated at least 10 meters.

Table 3. Quality standards for certified and commercial seed, Honduras, 2002.

 

 

 

 

 

 

 

 

 

 

 

Category Certified Seed Commercial Seed
Seed purity a 2 98 % 97 %
Seed of other varieties s 3 seeds/kg 10 seeds/kg

. Seed of other crops 5 2 seeds/kg 4 seeds/kg
Seed of weeds 5 2 seeds/kg 4 seeds/kg
Inert material 5 3% 3%

, Germination 2 80% 80%
Humidity 5 14% 14%
Seed with holes 0% 3%

 

 

Source: CERTISEM

a The percent of seed that has a cover even if the seed is rotten, wrinkled and has holes.

 

In addition to the field specifications, the bean seed crop must meet specific field

tolerance standards. The field tolerances for each seed category12 are shown in Table 4.

 

'2 CERTISEM does not specify seed tolerances for breeders seed, since the breeder is

responsible for insuring its quality.

50

 

Table 4. Field tolerances for different categories of bean seed, Honduras, 2002.

 

 

 

 

 

 

 

 

 

 

 

Presence of: Foundation Seed Registered Seed Certified Seed
Other varieties 0% 0.1% 0.2%
Other crops 0% 0% 0%
BGYMVa 0.5% 1% 1.5%
Antrachnose 0.5% 1% 1.5%
Bacterial blight 0.1% 0.3% 0.5%

3 Bean Golden Yellow Mosaic Virus

Source: CERTISEM

 

 

CERTISEM reported that plots are most commonly rejected because of variety
impurity, which is detected by CERTISEM staff when they visit the fields during the
development and flowering stage. On very few occasions, plots are rejected due to the
presence of diseases.

According to key informants, CERTISEM staff face two main problems. First, its
trained technicians periodically leave the institution following a change in the
government. Second, the agency has too few vehicles and personnel to inspect seed plots.
Thus, when there is a high demand for seed inspections, staff have to make arrangements
with the producers to pick up the technicians and drive them to the seed plots.
Consequently due to staff shortages, sometimes it takes CERTISEM more than 14 days to
carry out seed germination tests and return the results to their clients.

While Honduras has not yet approved intellectual property rights protocols for
seed, a draft law has been submitted to Congress for its approval. Thus, presently,

anyone can multiply any variety without paying royalties.

51

4.2.2 Breeder, Foundation, and Registered Bean Seed Producers

Interviews with key informants revealed that Zamorano and “The Directorate of
Research in Science and Agricultural Technology” (DICTA) are the only organizations
that produce breeder seed, foundation, registered, and certified seed. The following
sections describe information collected from key informants interviewed at Zamorano and
DICTA.

4.2.2.1 Zamorano/PIF

Zamorano has two distinct units involved in bean seed production. “El Proyecto
de Investigaciones en Frijol” (PIF), a research unit that produces breeder, foundation and
registered seed, and the Zamoempresa de Cultivos Extensivos (ZCE), a unit of Zamorano
that produces certified seed.

4.2.2.1.] Production

PIF produces breeder, foundation, and registered seed (small red and small black
beans) for the Central American and Caribbean region on its own plots (i.e. does not
contract with farmers). While PIF has contributed most of the germplasm (50%) in
varieties it has released for use in Central America and the Caribbean region, material
from CIAT (20%), University of Puerto Rico (20%), and DICTA (10%) have also been
used in making these crosses (Rosas, J .C., 2002, personal communication).

PIF maintains a seed bank which includes landraces breeding lines and varieties
released in Honduras (i.e. Tio Canela, Dorado, Catrachita, Danli 46, DICTA 113, DICTA
122, Don Silvio, Acacias 4, Esperanza 4, Zamorano, Desarrural). In addition, PIF

maintains unreleased germplasm (i.e. Milenio, Yeguare, and Don Victor), which were not

52

released due to some problems (e. g. discoloration of the grain from excessive water).
4.2.2.1.2 Quality Control
PIF’s is responsible for insuring the quality of its seed. No government agency
inspects its seed because they have that license to produce their own seed and maintaining
its reputation provides sufficient incentive to insure good quality seed.
4.2.2.1.3 Distribution and Sales
In recent years, PIF has produced about 1.8 mt of foundation/registered seed per
year. PIF sells registered seed to ZCE and various agricultural related projects. In
addition, PIF assists projects to which it provide seed to establish demonstration plots so
farmers can observe and compare different varieties. In the future, PIF plans to expand
its collaboration with the national program, DICTA (Rosas, J .C., 2002, personal
communication).
4.2.2.1.4 Prices
The price at which PIF sells its foundation seed is for L 2,000 per bag of 45 kg.
(Rosas J.C., 2002, personal communication).
4.2.2.2. DICTA
4.2.2.2.1 Production
DICTA produce breeder seed13 of its own varieties and has permission from
Zamorano to produce foundation, registered, and certified seed of the same small red
varieties that PIF produces. DICTA produces foundation seed on its fields in Comayagua

(i.e. they do not contract), although it sometimes obtains foundation seed fiom PIF. In

 

'3 CIAT has collaborated with the National Bean Program from 1989 to 1993 and with
DICTA since its creation in 1993.

53

recent years, DICTA has released several varieties (e.g. DICTA 113 and DICTA 122) and
has multiplied varieties bred by PIF (e.g. Tio Canela, Don Silvio and Dorado).
4.2.2.2.2 Quality Control

DICTA staff are responsible for insuring the quality of its foundation seed.
However, for registered and certified seed, DICTA must register the plot with
CERTISEM and pay for the seed label. Although not a major problem, the most common
reason that DICTA’s seed is rejected is due to weeds in the field.

According to DICTA personnel, because DICTA is understaffed, it is only able to
manage up to 7 has of seed of all types in its production fields during each growing
season.

4.2.2.2.3 Distribution and Sales

In the past, DICTA has sold registered seed to Hondugenet —a private seed
company. However, in recent years (2000-2002) DICTA only sold seed to Hondugenet
once, since after 2002 Hondugenet began to produce commercial seed instead of certified
seed. In addition, DICTA occasionally sells foundation seed to NGOs . While DICTA
sells seed of all types to farmers, NGOs, and agricultural related projects, DICTA is a
relatively minor actor in the seed market, as selling only 181 kg, 1.6 mt, and 2.3 mt of
foundation, registered, and certified seed combined in 1999, 2000, and 2001, respectively.
All sales were in Danli. The price at which DICTA sells seed varies by seed category. In
2002, DICTA sold the 45 kg bags of foundation, registered, and certified seed for L

2,000, L 1,800, and L 1,500, respectively (Escoto, 2002, personal communication).

54

4.2.3 Certified and Commercial Bean Seed Producers
Key informants reported that Zamorano (ZCE) and Hondugenet are the major
certified and commercial bean seed producers and sellers in Honduras. Zamorano only
sells certified seed and Hondugenet sells both certified and commercial seed. Additional
minor seed producers include DICTA14 (certified seed), as well as some NGOs
(commercial seed). The following sections describe information collected from key
informants from ZCE, bean seed contractees, and Hondugenet.
4.2.3.1 Zamorano/Zamoempresa de Cultivos Extensivos
ZCE is in charge of producing and marketing certified seed. It obtains registered
seed from PIF (70%) and DICTA (30%, mostly Dorado).
4.2.3.1.1 Production
ZCE produces about 45,455kg to 54,545 kg of certified seed per year. About 50%
of the seed is produced by ZCE staff at Zamorano and about 50% is produced by
contractors at San Juan de Flores. In 2001, Tio Canela (75%) and Dorado(25%) were the
varieties produced by ZCE. In the future, ZCE speculates they will produce mostly
Amadeus 77, due to its desired light red color.
4.2.3.1.2 Quality
To insure that contractees produce high quality seed, ZCE requires its contractees
to produce seed under irrigation during the dry season (J an-Apr), which reduces

uncertainty and disease problems. In addition, ZCE requires its contractees to follow its

 

” Currently DICTA sells small amounts of certified seed to NGO’s (e.g. CRS,
Movimundo, and World Neighbors), institutions (e.g. F AO, ICADE, and Zamorano), and
farmers.

55

recommendation regarding plant spacing, weeding, disease and insect control, and
roguing. Furthermore, ZCE staff visit the farmers to select the production fields, monitor
land preparation, planting, crop development, flowering and harvest.

4.2.3.1.3 Sales

Before 2002, ZCE sold seed mainly in 50 pound bags (22.7 kg) and in a few cases
in 10 (4.54 kg) and 25 (11.36 kg) pound bags . However, starting in 2002, ZCE began to
sell seed in 44 lb (20 kg) bags.

In 2001 ZCE distributed all seed it produced from its office at Zamorano. While
ZCE made DUWEST (a national distributor of Dupont’s agricultural chemicals) an
authorized distributor of its certified seed by early 2002, by the 2002 primera planting
season, ZCE had not yet started to distribute seed through DUWEST. DUWEST
distributes agricultural inputs throughout Honduras.

Key informants at ZCE reported that they have not had any problems getting its
seed inspected or certified, although sometimes CERTISEM employees did not carry out
all of the scheduled visits.

ZCE sold certified seed for L 14/lb (L 30.8/kg) in 2001 and plans to sell for L
15.9/1b (L 35.0/kg) in 2002. ZCE staff would not give the price at which they plan to sell
certified seed to DUWEST because they said it was classified information.

4.2.3.1.4 Contract Terms

ZCE started contracting farmers to produce certified seed after Hurricane Mitch

(1999), when the demand for seed increased dramatically. In selecting contractees, ZCE

looks for educated and trustful farmers «farmers who they already know by reputation.

56

In addition, they look for contractees who are located near Zamorano and have access to
capital and irrigation. Since 1999, ZCE has used contractees located in San Juan de
Flores, Francisco Morazan (38 km from Zamorano).

ZCE makes a verbal contract with farmers which specify the area to be planted, an
expected yield, and delivery place and date. ZCE agrees to pay the land registration fee"
(which prohibits farmers from legally selling the seed to anyone except ZCE) and
provides the contractee with registered seed. The contractee agree to pay for all inputs,
except seed, and agrees to host regular inspections by ZCE and the required inspections
by CERTISEM staff. At harvest, ZCE collects the seed at the farm where it is grown,
pays the transport costs from the farm to Zamorano and returns rejected grain to the
farmer, unless the farmer agrees to accept a lower price (i.e. grain price) for the rejected
seed. ZCE pays contractees L 1.5/lb (L 3.3/kg) above the SIMPAH" bean price, although
the price premium can be renegotiated if the grain price increase to a high level.

The ZCE has not had any problems with its contractees (e. g. rejection of bean
seed fields, legal disputes). However, some farmers expressed concern about the amount
of grain that ZCE accepts as seed. Rather than having their seed rejected, contractees
would prefer to receive a price decrease (i.e. from L 650/qq to L 550/qq, but not lower (1

qq = 100 lbs)), as long as all grain is accepted.

 

‘5 Along with the land inscription, ZCE pays for all visits by CERTISEM.

“5 SIMPAH is a government organization that monitors and reports the prices of major
agricultural commodities grown in Honduras.

57

4.2.3.1.5 Contractee’s concerns

While the contractees were generally satisfied with the above described contract,
some contractees are unhappy with the percentage of their production that was rejected as
seed (Table 5). The contractee reported that while Hondugenet paid farmers only L 5.5/lb
(L 12.1/kg) of seed in 2002, it did not reject any seed. So the farmer argued that it was
more attractive to sell to Hondugenet than to ZCE which paid a higher price but also had
a higher rejection rate. However, ZCE need to maintain high quality standards to protect
against moral hazard, since it markets certified seed while Hondugenet only markets
commercial seed.

Table 5. Percentage of seed rejection of seed delivered to the Zamoempresa de
Cultivos Extensivos by a representative contractee, Honduras, 2002.

 

 

 

 

 

 

 

 

 

 

 

 

 

Year Delivered Amount of Sales as:
(crq)a
Seed grain percent sold

as seed
2001-2002 208 158 50 75.96%
2000-2001 105 72 33 68.57%
1999-2000 230 1 50 80 65.22%
Mean 181 126.66 54.33 69.92
a qu = 100 lbs.
Source: Key informant interview, 2002.

 

Second, contractees expressed concern about the lack of transparency in ZCE’s seed
evaluation process. After transporting the seed to Zamorano, as part of processing, ZCE uses
a screen to select out small size kernels. However, the contract does not specify the criteria
for rejecting seed (e.g. grain size, foreign matter, germination rate). Thus,
to avoid future problems with contractees, ZCE should specify in the contract, the size of the

screen that it uses to select seed and other criteria for rejecting seed.

58

Third, some contractees expressed concern that the price that ZCE paid for seed has
declined. In 1999, contractees received a price of L 7/lb (L 15.4/kg). This high price was
paid because ZCE was certain that it would sell a large quantity of seed at a high price to the
Post-Mitch alleviation projects. However, in 2001 and 2002, ZCE paid its contractees L
5.5/lb (L 12.1/kg) and L 6.5/lb (L 14.3/kg), respectively.

In addition, contractees expressed concern about the difficulty of obtaining loans
from banks. Typically banks require borrowers to provide urban collateral as a guarantee
against default. In addition, the banks value these assets at 50% of their market value and
then issue a loan for 50% of that value.

Finally the contractee reported that to successfully produce bean seed, farmers must
have access to an irrigation system because the seed crop must be grown during the dry
season.

4.2.3.2 Hondugenet

In the past, Hondugenet obtained its foundation seed from DICTA. However, the
firm had not purchased seed from DICTA17 since 1998. Currently rather than buying
foundation or registered seed, Hondugenet produces seed from its own seed stocks. Thus,
although Hondugenet can only sell commercial seed, it is able to sell its commercial seed at
the same price as certified seed due to its good reputation.

4.2.3.2.1 Seed Production
Hondugenet produces seed in Francisco Morazan and El Paraiso because those are

the country’s most productive bean areas and Hondugenet has a long relationship with its

 

'7 Hondugenet has also bought certified seed from Zamorano to produce their commercial
seed.

59

contractees in these areas. About 70% of Hondugenet’ 5 seed is produced by contractees (6-7
farmers), and the firm produces 30% in leased fields.
4.2.3.2.2 Contract terms

According to key informants at Hondugenet, good contractees are farmers who do not
place the crop at risk for trying to cut costs, and who work on a fixed schedule. Hondugenet
selects farmers who have farms on which beans can be grown isolated from other crops, are
easy to access, have an irrigation system and flat fields, own or can rent a tractor, have
sufficient labor, are trained in crop management (e.g. herbicides, insecticides), and have
fields close to the processing plant. Also, to insure the contractees produce high quality seed,
Hondugenet staff makes 8-10 inspections to contractees’ fields.

Even though there is no written contract, Hondugenet’s requires that the contractees’
seed meet the standards required by law to be considered commercial seed. The informal
contract specifies the area to be planted, that the contractee delivers18 the seed to the
processing plant in “Ojo de Agua”, and that Hondugenet will pay for the grain at delivery
(14% humidity). All input costs are incurred by the contractee, except for seed which is
provided by Hondugenet. Similar to ZCE, Hondugenet pays a given percentage (not given,
confidential) above the grain market price given by SIMPAH.

Key informants at Hondugenet reported no problems with their contractees and
rejected almost no seed. Also, CERTISEM has not rejected any seed produced by

Hondugenet’s contractees and CERTISEM staff have inspected the seed in a timely manner.

 

'8 Hondugenet pays CERTISEM just for sampling and for the label (i.e. field inspections
are not done), because they only sell commercial seed.

60

4.2.3.2.3 Sales

Hondugenet sold approximately 18 mt of commercial bean seed in 1997 and 1998,
and 136 mt in 1999, 2000, and 2001. Hondugenet sells its bean seed to large-scale farmers,
distributors, NGOs, and projects (e.g. CARITAS, CRS, CARE, FAO, GTZ, Movimundo,
PNUD, and the Red Cross)—mostly in Tegucigalpa (80%) but also in San Pedro Sula (20%).
Hondugenet produces several varieties. However in 2001, it mainly sold Tio Canela (68 mt),
Dorado (45 mt), and Dicta 113 (14 mt).

Hondugenet does not have any authorized distributors. Rather, it sells seed primarily
to 8-10 customers who buy large quantities which they redistribute. These buyers receive
a discount of L l/lb ( L 2.2/kg) below the retail selling price. In 2001, Hondugenet sold its
seed in Tegucigalpa and San Pedro Sula for L 14/lb (L 30.8/kg) and planned to sell
commercial seed in 2002 for L 15/lb (L 33/kg). Also, Hondugenet does not sell seed on
consignment because returned seed might have a low germination rate.

Hondugenet primarily sells its seed in 50 lb (22.7 kg) bags because this size is easy
to handle and it is the amount needed to plant one manzana (0.7 Ha). However, they will sell
seed in a 20 lb (9 kg) bag to a farmer who asks for it.

4.2.3.2.4 Hondugenet Bean Seed Production Role

Hondugenet staff reported that bean seed sales is a side business, rather a primary part
of their seed business (i.e. produce bean seed just to give a service, not interested in
participating in seed scheme). In addition, they commented that the bean seed business is
tied to development of a new variety—demand is greatest afier a new variety is released and

then declines. Furthermore, they only sell bean seed in Tegucigalpa and San Pedro Sula

61

because farmers in other areas/departments have low purchasing power and typically buy
seed from their neighbors.

Finally, Hondugenet staff were pessimistic about the sustainability of artisan seed
projects and providing seed vouchers to NGO-beneficiary farmers. They asserted that
farmers sell artisan seed at grain prices out of necessity (i.e. they sell the seed at the grain
price because local farmers are not willing to pay a premium for seed) and using vouchers
to sell seed to farmers will not work because of corruption.

4.2.4 NGOs and Artisan Seed Producers

4.2.4.1 NGOs

Over the past 20 years, artisan seed production has been promoted by the Ministry
of Agriculture and NGOs with support from Zamorano and CIAT. Key informants at CIAT,
reported that under the project “Production and Commercialization of Improved Artisan
Seed”(PCIAS‘9), 10 NGOs (i.e. APS, CCD, World Neighbors, CONSULUPE, Grupo Guia,
MOVIMONDO, ICADE, and CISP) are assisting small-scale farmers to produce commercial
seed of beans, maize, sorghum, soybean, rice, and peanuts. The project covered the
departments of Colon, Santa Barbara, Ocotepeque, El Paraiso, and Francisco Morazan.

The NGOs purchased registered seed from either Zamorano or DICTA and then
distributed it to their farmers. In the first year, the NGOs gave the inputs to the farmers, but

in the second year the farmers had to pay for the inputs. The NGOs do not use contracts,

 

'9 PCIAS is very similar as other artisan seed schemes and to the scheme proposed in this
research. The major differences between PCIAS and the proposed seed scheme are that
under PCIAS farmers sell their seed in the region where it is produced, while the goal of
the proposed is to contract a private input supplier (DUWEST) and market the seed
through its distribution network.

62

since the farmers were considered to be beneficiaries. In 2001, farmers from World
Neighbors sold bean seed in 25 lb (1 1.36 kg) bags at L 8/lb (L 17.6/kg). According to CIAT
staff bean seed production costs averaged around L 6/lb (L 13.2/kg).

World Neighbors (WN) and Catholic Relief Services (CRS) are two active bean seed
distributing NGOs. Also, some projects like PRODERCO, were involved in bean seed
distribution. These NGOs and PRODERCO distributed seed mainly in 25 lb (11.36 kg)
bags. These project and NGOs gave the bean seed at highly subsidized rates. However,
CRS was willing to use a voucher system.

4.2.4.2 Role of CIAT

CIAT has played a major role in supporting PCIAS, including selecting leader
farmers in each village to help manage the seed production scheme, training the farmers in
seed production, and assisting the farmers to sell the seed that they produce.

CIAT’s staff reported that farmers in some regions had low yields or the crop failed
due to low rainfall, since the farmers do not have irrigation systems.

4.2.4.3 ARSAGRO

ICADE, one of the NGOs that promotes artisan seed production under PCIAS,
works with several farmer groups including ARSAGRO--farmers association that offers
agricultural services to its members. ARSAGRO staff and farmers were interviewed to gain
insights from participants regarding how artisan schemes function. To become a member,
farmers must pay a fee, agree to participate in meetings, and own land. The farmers selected

to produce commercial seed had previously taken seed production training”. Farmers

 

2° From 1990-1992, The Ministry of Natural Resources trained farmers in artisan seed
production.

63

reported that the advantages of being associated with ARSAGRO is that membership gives
them access to inputs, services (i.e. plowing), and credit.

ARSAGRO’ 5 members produces seed in the municipalities (department subdivisions)
of Danli, El Barro, Arauli, and San Matias. In 2001, ARSAGRO farmers produced
commercial seed of Dorado, Tio Canela, DICTA 113, Amadeus-77, and Carrizalito
(Amadeus and Carrizalito just as trials since they were not released yet). While ARSAGRO
does not require its seed growers to sign written contracts, farmers must agree to grow seed
in a field that is acceptable to ARSAGRO and deliver it to ARSAGRO at harvest. In 2001,
ARSAGRO paid farmers L 5/lb (L l 1/kg) for dry (14% humidity) unselected seed in bags,
delivered to ARSAGRO’S office in Danli. In 2002, farmers were negotiating to receive L
6/lb (L 13.2/kg) of dry and selected seed. ARSAGRO markets its seed in the 25 lb (11.36
kg) bags (provided by CIAT) to farmers in the region, since it does not have distributors to
market the seed.

A summary of the agents and functions of seed suppliers in Honduras is presented

in Table 6.

64

Table 6. Seed Production and marketing in Honduras, 2002.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Activity / Participants Description
Breeding
Zamorano/PIF makes crosses
DICTA makes crosses
Seed Multiplication
Zamorano
PIF multiplies breeder seed up to registered seed
ZCE multiplies registered seed to produce certified seed
ZCE’s contractees multiply registered seed to produce certified seed
DICTA multiplies breeder seed up to certified seed
Hondugenet
Firm Hondugenet multiplies certified seed to produce
commercial seed
Hondugenet’s Hondugenet’s contractees multiply certified seed to produce
contractees commercial seed
NCO?
NGOs central office collaborate with CIAT in its seed multiplication scheme;
NGOs multiply certified seed to produce commercial seed
NGO Farmers participate with NGOs (i.e. ARSAGRO farmers) to

multiply certified seed to produce commercial seed

 

Seed Inspection

 

CERTISEM

inspects all fields used to produce certified seed and all
certified and commercial seed

 

Seed sales/Distributionb

 

 

 

 

Zamorano/ZCE sells seed at Zamorano’s campus
Hondugenet sells its own seed in Tegucigalpa and San Pedro Sula
NGOs helps farmers sell their seed in neighboring towns

 

 

a NGOs (i.e. APS, CCD, World Neighbors, Consulupe, Grupo Guia, Movimundo,

Lcade, CISP).

DUWEST made a contract to distribute ZCE’s seed but had not started by June 2002.
Source: Key informants, 2002.

 

65

 

4.2.4.4 Potential Bean Seed Producers

In addition to the ARSAGRO farmers, a group of farmers from the north coast were
interviewed to see if they were interested in participating in the proposed bean seed
production scheme. This farmer group were contacted because they were experienced bean
farmers and had access to extension services through the National University of Honduras.
In addition, the north coast farmers can grow beans during December to February. Thus,
they can produce bean seed for sale in the primera season without irrigation. Although they
had never produced bean seed, this farmer group were interested in leaming how to produce
seed and participate in the proposed scheme.

4.2.5 Input Distributor

4.2.5.1 DUWEST

While several firms (e.g. CADELGA, FERTICA, BAYER) distribute agricultural
inputs in Honduras, DUWEST was the only firm that expressed an interest in distributing
bean seed. DUWEST has exclusive and non-exclusive distributors/agents who are based in
the main department cities of Olancho, Yoro, El Paraiso. Also, in addition to department
capitals, DUWEST sells in fairly big cities. DUWEST markets its products through
subdistributors (e. g. REISA and PROTECNO). For example, DUWEST has around 10-15
subdistributors/agents in Danli, El Paraiso, including Alagro and Agropecuaria Rodas.
DUWEST sells its inputs on a cash basis. These agents markup products 10-15% above the
price they paid to DUWEST.

In 2002, DUWEST planned to sell bean seed to subdistributors in El Paraiso and

Olancho for the postrera season at L 17.27/lb (L 38/kg). DUWEST expected to obtain its

66

seed from ZCE for L 13.82/lb (L 30.4/kg) of bean seed. The price of L 17.27/1b (L 38/kg)
in Danli, includes the 20-25% markup (2-3% is to cover transportation costs) from the price
DUWEST paid to ZCE. Transportation costs (calculated as 17.27*0.02-17.27*0.03) are L
0.35 - L 0.52 per lb (L 0.76-L 1.14 per kg). The final price to the farmer would be around a
range of L 19/lb (L 41.8/kg) and L 19.86/lb (L 43.7/kg) with the markup of 10-15% from
the subdistributor.

4.2.6 Agricultural Banks

Staff of four banks were interviewed: Banco de Occidente (BO), BAN ADESA, BGA,
and BAMER. Banco de Occidente (BO), BANADESA, and BAMER have given loans to
bean farmers for up to three years before 2002.

After hurricane Mitch, BGA reduced the amount of their direct loans to agriculture
and restructured its loans. In 2000, BGA stopped making loans to most bean farmers
because of the risk involved in agriculture. In addition, the bank shificd their lending to
more profitable crops and lengthened the repayment time of their late loans.

On the other hand, Banco de Occidente has directed more funds to small farmers,
using both its own funds and discounted funds (i.e., subsidized interest rate) that is obtained
from the government.

BAMER and BAN ADESA continue to extend loans to bean farmers, except that now
they are also extending low interest rate loans with discounted funds recently made available

by the government.

67

The value of loans to bean farmers accounts for 2-3% of the agricultural account in
some banks like BO but less than 1% in BGA. As Table 7 shows, the value of loans

extended to farmers by each bank ranges from L300,000 to L7,000,000.

4.2.6.1 Recent Lending Activities

Table 7. Value of loans extended to bean farmers by several banks in Honduras, 2002.

 

 

 

 

 

 

 

 

 

 

 

Total Value of Loans (In Leeriras)
Year 2002 2001 2000 1999
Bank
BAMER NA 7,000,000 4,500,000 3,000,000
BAN ADESA 3 ,100,000 3,900,000 4,300,000 6,000,000
Banco de NA 300,000,000a NA NA
Occidente
BGA 300,000 NA NA NA

 

Source: Key informant interviews, 2002.

 

$1 = L 16.44

 

NA= Not available
a Total amount in the agricultural account for that year.

 

Banco de Occidente, BANADESA, and BAMER extends minimum loans of L5,000,
L 3,000, and L10,000, respectively. BGA lends 75% of production costs to clients with

annual sales greater than $120,000, which eliminates small scale farmers.

In the interest of reducing transaction costs, staff from the banks were asked if they
had agreements with rural cashiers“. BAMER and BGA have no agreements with rural
cashiers. On the other hand, Banco de Occidente and BANADESA manages accounts with

rural cashiers and extends loans to them. This not only reduces transaction costs to the bank,

 

4.2.6.2 Banking Operations

2‘ Rural cashiers are entities that work in collaboration with banks to facilitate loans to
farmers in the rural areas.

68

 

but also to the farmer which can make it easier to farmers to invest in bean seed production.
In addition to having agreements with rural cashiers, BAN ADESA uses government funds
to make loans to farmers who do not have collaterals.

While BGA made it clear that it is not willing to extend loans to bean seed farmers
because of the risk involved and late payments, BAN ADESA and Banco de Occidente are
interested in working with small-scale bean farmers.

4.2.6.3 Loan Conditions
4.2.6.3.1 Interest Rate

BGA charges an interest rate of 26-32% depending on the client. When BGA uses
FONAPROVI22 funds, it changes 10 %. However, BGA has not extended any loan to bean
farmers with FONAPROVI funds because the bank bears the risk. BAMER used to charge
a 19% interest rate, but with FONAPROVI funds they lend at 10%. Banco de Occidente
used to charge an interest rate of 22-28% with their funds and 17.22% with FONAPROVI
funds. However in 2002, it charged 10% after legislation lowered the F ONAPROVI firnds
interest rate. BANADESA charges 10% interest rate on FONAPROVI funds, and 24% on
its own funds.

4.2.6.3.2 Time Horizon

BGA requires farmers to repay their loans in six to nine months. BAMER requires

farmers to repay their loans in nine months, but allows farmers 18 months for FONAPROVI

funds. Farmers can opt to repay their loans in two payments instead of one. BO requires

 

22 F ONAPROVI is a financial institution that provides loans for production of goods and
housing for lower income groups.

69

farmers to repay their working capital loan in nine months and fixed asset loans in five years.
BANADESA requires farmers to repay any kind of loan in six months.
4.2.6.3.3 Collateral

BGA requires farmers to have a collateral, usually an urban mortgage. It allows
farmers to borrow up to 60% of the bank’s appraised value of the mortgage, or if it’s a
pledge, farmers can borrow up to 40% of the appraised value and the pledge has to be insured
and turned over to the bank. BAMER accepts both mortgage collaterals and pledges. It
allows farmers to borrow up to 60% of the bank’s appraised value of the mortgage. Banco
de Occidente accepts fiduciary (up to L 25,000 and two guarantors), mortgage (up to
L3,000,000, 60% of the appraised value) or a pledge (40—50 % of appraised value).
BAN ADESA accepts fiduciary (with guarantors), mortgage (60% of the appraised value) or
pledge (40-50 % of land registry appraised value) as a collateral.

4.2.6.3.4 Group Loans

BGA is willing to extend loans to individual farmers or to groups of farmers if a
recognized institution agrees to serve as a collateral and provides technical assistance to these
farmers. BAMER is willing to extend loans to individual farmers or to groups of farmers.
If its in a group, they would deal only with the president of the group. Banco de Occidente
prefers to extend loans to individual farmers. If loans were made to a group of farmers, the
group would all be responsible for repaying the loan. BAN ADESA is willing to give loans
to individuals or to groups of farmers. BANADESA is willing to work with bean farmers,

as long they meet the following requirements: photocopy of ID, municipal solvency,

70

complete ownership of collateral, up to date with taxes, 2-3 bank references; for a group:
authorized act, legal status and CNC and COCOCH proof (legal registration).
4.2.6.3.5 Interest in Loaning to Scheme Farmers

Banco de Occidente, BAMER and BANADESA were willing to provide credit to
small farmers. Nevertheless, BAMER’s minimum loan is two to three times larger than
BO’s and BANADESA’s loan. Thus, many small-scale farmers might not be interested in
obtaining a loan from BAMER, if they only wanted working capital to cover the cost of
inputs like seed, fertilizer and pesticides which would only cost approximately L 5000/ha.
Even though BANADESA and Banco de Occidente are willing to lend to small-scale
farmers, small scale farmers may not be willing to ask for a loan. Given the production risks
involved (i.e. rainfall) farmers might be hesitant to take a loan because they fear losing the
land or other collateral. In addition, small-scale farmers might be hesitant to take a loan due
to transaction costs involved (i.e. paperwork).

Finally, BGA is not willing to work with small-scale farmers because they frequently
do not repay loans on schedule and due to the risk involved in lending to small-scale farmers.
4.3 Summary

Bean seed supply is characterized by a series of seed production stages up to the
distribution of seed through a marketing network. The seed production stages are the
development of breeder seed and then a series of multiplications from breeder seed to
foundation seed to registered seed to certified seed and finally commercial seed. This

process is supervised by a governmental agency, CERTISEM.

71

PIF and DICTA are the only two entities that produce breeder, foundation, and
registered seed. PIF sells its registered seed to ZCE and assists NGO and agricultural related
projects with bean seed to establish demonstration plots for farmers.

ZCE is the major certified bean seed producer in Honduras and Hondugenet is the
major commercial bean seed producer. While ZCE produces only certified seed, Hondugenet
produces both certified and commercial seed, the former being produced erratically. Another
minor participant that produce certified seed is DICTA and some NGOs produce commercial
seed.

ZCE contracts with farmers to produce certified seed. Although not a major problem,
some contractees have expressed concern on the lack of transparency of the contract,
specifically the screen size used to reject grain and the seed price paid to them. Hondugenet
also, contracts with farmers to produce its commercial seed. The major differences between
the contracts is the delivery place and the rejection rate of the grain. The formula price is
unknown for both contracts .

In 2001 and 2002, some NGOs together with CIAT, trained small-scale farmers to
produce commercial seed in several regions of Honduras. The main differences between
CIAT’s scheme and the one proposed in this study is the larger scope of the distribution and
the inclusion of a private company to do the distribution.

DUWEST was the only major agricultural input distributor in Honduras interested
in distributing bean seed as part of the proposed scheme. The final retail price in Danli of
ZCE-produced and DUWEST-distributed in 2002 was around L 19/lb (L 41.8/kg) - L

19.86/1b (L 43.7/kg).

72

Even though banks have played a minor role in small-scale farmers bean production,
Banco de Occidente and BANADESA seem promising for small-scale farmers acquiring
capital to invest for the bean seed production scheme, mainly because of their disposition

towards small-scale farmers.

73

Chapter V.
5. Bean Seed Demand In Honduras

To assess the demand for improved bean seed, farmers were surveyed in two
bean-producing regions of Honduras to characterize their farming system and seed use.
These data are used to assess the potential demand for seed of recently-released modern
bean varieties and implications for the proposed seed scheme.
5.1. Farmer’s Characteristics, Seed Use, Production, and Sales

The survey was carried out in El Paraiso (N=36) and Olancho (N=36).

5.1.1. Socioeconomic Characteristics of the Farmers

The sampled farmers owned an average of 5.37 ha of land. However, this does
not mean that they all had a title for the land, but that they had been farming the land for
several years and no one had challenged their use of the land. Almost half of the farmers
(47%) reported having land with poor to normal soil quality. The average age of the
farmers was 51 years, ranging from 22 years to 82 years. The farmers had few years of
formal education. On average, the farmers had completed the fourth grade, but their
education level ranged from no school to university (16 years).

5.1.2. Bean Area

Corn is the most—important staple grain in Honduras, followed by beans. Results
from the survey indicates that 86% and 57% of bean farmers grew corn in the primera
(2002) and postrera (2001) seasons, respectively. In addition, 35% of bean farmers grew
corn in one season, 54% grew it in both seasons, and 11% did not grow com at all. The

average area planted to beans in El Paraiso and Olancho were similar in the postrera (2.12

74

ha and 2.02 ha, respectively), but El Paraiso planted more in the primera (1.37ha and 0.76

ha, respectively)(T able 8).

Table 8. Average area and average of farmers planting corn and bean in postrera
(2001) and primera (2002), by department, Honduras.

 

Average area (ha)

Average of farmers planting crop

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(ha)
Department El Paraiso Olancho El Paraiso Olancho
Season Bean Corn Bean Corn Bean Corn Bean Corn
Postrera 2.12 1.16 2.02 1.81 2.12 2.45 2.02 2.72
N 36 36 36 36 36 17 36 24
Primera 1.37 2.37 0.76 2.93 1.70 2.59 1.04 3.63
N 36 36 36 36 29 33 26 29

 

 

Source: Farmer survey, 2002.

 

5.1.3. Source of Bean Seed

Farmers obtained bean seed from several different sources (Table 9), which are

predominantly associated to the “traditional stage” of seed development. In both

departments, the two most important seed sources were saved seed from the previous

harvests and seed obtained from another farmer. In El Paraiso and Olancho, most of the

bean area was planted with saved seed (73%, 63%) or seed obtained from another farmer

(20%, 26%). Farmers reported that they usually traded or bought seed from another

farmer when they ran out of seed, wanted to replace their stock with fresh seed, or wanted

to change varieties. When farmers purchased seed from another farmer, they paid a price

close to the price of grain.

75

 

Table 9. Hectares farmers planted (percentage) to bean by seed sources and
department, during four seasons (postrera 2000 to primera 2002), Honduras.

 

 

 

 

 

 

 

Department (percentage) Totala
Seed Source El Paraisob Olanchoc
(Ha = 246.35) (Ha = 245.78)

Saved from previous harvest 73.4 63.0 68.2%
Another farmer 20.0 25.9 22.9%
NGO/Project 4.0 7.8 5.9%
Purchased in grain market 1.7 1.6 1.7%
Certified seed dealer 0.9 1.7 1.3%

 

 

 

 

a Approximation. Exact percentage can be obtained by using percentages in each
gepartrnent and their total hectares.

N=36 °N=36
Source: Farmer survey, 2002.

 

 

 

In recent years, NGOs and Projects have been an important source of seed for
small farmers who have faced a seed shortage due to droughts and hurricanes. Two
examples were the Zamorano Post-Mitch projects in 1999 and 2000-2001. While only 4%
and 8% of the area in El Paraiso and Olancho, respectively, was planted with seed
obtained from an NGO or Project, 21% of the farmers reported previously receiving a
modern variety from an NGO or Project. Finally, only a small percentage of the bean
area in El Paraiso (1%) and Olancho (2%) was planted with seed obtained from a
certified dealer. Furthermore, although 28% of the farmers had purchased improved seed
from a certified dealer at least once, 72% of the farmers had never purchased improved

seed from a certified dealer.

76

5.1.3.1 Reasons

Farmers gave several different reasons for not buying certified seed (Table 10).
The most common reason was that certified seed is too expensive (El Paraiso, 52.4% and
Olancho, 72%). This reason can be interpreted in several ways, including that the farmer
did not have the capital available to buy the seed, or that he/she did not believe that
improved seed was worth the extra investment, given the extra revenue that it will
generate. The second most important reason farmers gave for not buying improved seed
was that they can produce their own seed. While the seed that farmers produce is of
lower quality than certified seed, it is a good substitute when farmers cannot afford to
purchase certified seed. The third most important reason that farmers reported was that
certified seed was not available in the towns where they live (El Paraiso,14% and
Olancho, 6%). Because certified seed dealers usually only sell seed in major cities,
farmers must travel to a city to purchase it, which increases the cost of improved seed.
Contrary to expectation, very few farmers reported bag size as a reason for not purchasing

certified seed.

Table 10. Percentage of farmers quoting different reasons for not ever buying certified
seed, by department, Honduras, 2002.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Department Reasons
Too Not Bag size Produces Custom Likes
expensive available own seed landraces
El Paraiso 52.4 14.3 4.8 23.8 0.0 4.8
(N = 21)
Olancho 72.2 5.6 1.0 11.1 11.1 0.0
(N=18)
Total 62 10 3 18 5 3
13 missing (nonresponces). Source: Farmer survey, 2002.

 

77

5.1.3.2 Prices Paid
Compared to certified seed, seed that farmers purchase from another farmer is
relatively inexpensive. Farmers reported hypothetical prices based on what they think
they would pay when buying seed of traditional and recycled modern varieties from a

neighbor. The average of these reported hypothetical prices are shown in Table 11.

Table 11. Average hypothetical prices (Lempiras/Kg) for bean seed“l purchases among
neighboring farmers, by department and type of seed, primera 2002, Honduras.

 

 

 

 

Variety Department

El Paraisob (N = 46) Olancho (N = 41)
Traditional (N = 57) L 9.6 L 8.5
Modern (N = 30) L 8.6 L 7.7

 

 

 

a These are prices of traditional and recycled modern varieties and not resale price from
eed purchased from a certified dealer.
Traditional N=33, Modern N=13

Source: Farmer survey, 2002.

 

 

 

Bean seed was slightly more expensive in El Paraiso than in Olancho. Farmers in
El Paraiso reported that they would have to pay L 9.6/kg (L 4.36/lb) to a neighbor for
seed of a traditional variety and L 8.6/kg (L 3.91/lb) for a modern variety. In contrast,
farmers in Olanchopaid L 8.5/kg (L 3.86/lb) and L 7.7/kg (L 3.5/lb) for TVs and W5,
respectively. The small difference in the prices between modern and traditional varieties
is due to the price discount for MV (1'. e. darker red color). As expected, seed prices in
Olancho were lower than prices in El Paraiso because the surveyed villages in El Paraiso
were nearer to the department capital. Given that certified bean seed sold forL 35/kg (L
15.90/lb)(at Zamorano), it is clear why farmers with limited resources saved their seed or

bought seed from a neighbor instead of from a certified seed dealer. While the prices

78

farmers paid for seed obtained from another farmer was usually equal to the price of
grain, 36% of the farmers in El Paraiso and 9% of the farmers in Olancho reported that
they paid a higher price for seed than for grain.

5.1.4. Varieties Planted

Farmers planted both traditional and modern varieties. Farmers in the study area
planted a total of 22 different varieties over the four seasons (246 ha) from the postrera of
2000 to the primera of 2002. In El Paraiso, the farmers planted four different modern
varieties and in Olancho they planted five different modern varieties (Table 12). In the
other hand, in El Paraiso, the farmers planted six different traditional varieties and in
Olancho they planted 13 different traditional varieties.

In El Paraiso, Paraisito was the most widely planted variety in terms of area and
number of farmers planting it. On the other hand, in Olancho, Jutiquile was the most
widely planted variety in terms of area. Dorado and Tio Canela were the most widely

planted varieties in terms of number of farmers planting them.

79

Table 12. Area planted (ha) and number of farmers who planted the variety during any
of four seasons, (postrera 2000 to primera 2002), Honduras.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Department

El Paraiso Olancho
Variety Name Area Farmers Area Farmer Type
Dorado 58.5 14 60.2 14 Modern
Tio Canela 8.9 6 53.1 14 Modern
Zamorano 8.6 5 1.3 2 Modern
Catracho 0 0 1.4 1 Modern
Seda 0 0 0.7 1 Modern
Esteli 90 0.5 1 0 0 Modern
Danli 46 1 .4 1 l .2 2 Modern
Paraisito 1 50. 1 32 0.4 1 Traditional
Jutiquile 0 0 80.7 7 Traditional
Cuarenter’io 1 .9 2 ] 1 .3 9 Traditional
Chimino 0 0 9.1 4 Traditional
Payomo 8.1 3 0 0 Traditional
Vaina Blanca 0 0 7.5 6 Traditional
Cincuentefio 0 0 6.7 5 Traditional
Retinton 6.3 1 0 0 Traditional
Sangre de Toro 0 0 3.3 3 Traditional
Arbolito 0 0 2.8 1 Traditional
Nogue 0 0 2.7 2 Traditional
Marciano 2.1 2 0 0 Traditional
Chapin 0 0 l .8 1 Traditional
Tegucigalpa 0 0 0.9 1 Traditional
Rosita 0 0 0.8 3 Traditional
Total 246.4 67 245.9 77
Source: Farmer survey, 2002.

 

5.1.5. Traditional Varieties v.s. Modern Varieties

During the four seasons, farmers in El Paraiso planted a larger percent of their

bean area to traditional varieties (68%), compared to the Olancho farmers (52%).

Of the farmers who planted modern varieties from 2000 to 2002, some used

varieties released as early as 1984. However, most farmers had heard of the most

80

 

recently-released varieties« Dorado (1990) and Tio Canela (1996). In El Paraiso, all of
the farmers had heard of Tio Canela, but only 44% had planted it. On the other hand,
97% of the farmers had heard of Dorado and 85% had planted it. In Olancho, there was
less difference between farmers who had heard of each modern variety and who had
actually planted it. Almost all of the farmers (97%) had heard of both Tio Canela and
Dorado, 71% had planted Tio Canela, and 65% had planted Dorado. Dorado was planted
by more farmers because it was released six years before Tio Canela and therefore in a
later stage of adoption.

Although only a few farmers planted both a MV and a TV in the same season, the
farmers reported that they liked to try different types of varieties. For example, during the
period from postrera 2000 to primera 2002, a majority of the farmers in both departments

planted both a TV and a MV (Table 13).

Table 13. Percentage of farmers who planted modern varieties, traditional varieties,
and both traditional and modern varieties (not necessarily in the same season) during
four seasons (postrera 2000 to primera 2002), by department, Honduras.

 

 

 

 

 

 

 

 

 

Department
Type °f variety El Paraiso (N = 36) Olancho (N = 36)
Traditional only 30.6 30.6
Modem only 5 .6 16.7
Traditional and Modern 63.9 52.8
Source: Farmer survey, 2002.

 

 

5.1.6. Total Production
The farmers were asked to estimate the area that they planted to each variety and

the total harvest for each variety during four seasons (postrera 2000 to postrera 2001)

81

(Table 14). In El Paraiso, a larger percentage of the area was planted to traditional
varieties (65%) compared to Olancho (53%). Nevertheless, the total combined area
planted to traditional and modern varieties in Olancho is larger than in El Paraiso. In the
other hand, El Paraiso farmers had higher yields for both TVs (767 kg/ha) (1181 lb/mz)
and MVs (855 kg/ha) (1317 lb/mz) than farmers in Olancho (713 kg/ha (1098 lb/mz), and
44] kg/ha (679 lb/mz), for traditional and modern varieties, respectively). In El Paraiso,
MVs yielded higher than TVs. In contrast, in Olancho, MVs yielded lower than TVs. A
possible explanation to this unexpected result is what Mather (2003) found when
comparing yields of modern and traditional varieties, modern varieties were most widely
grown in areas where Bean Golden Yellow Mosaic Virus pressure was high and thus

traditional varieties could not be grown.

Table 14. Farmers total production and area by type of variety, postrera 2000 to
postrera 2001, by department, Honduras.

 

 

 

 

 

 

 

 

 

 

 

 

Department N Type of Production ha yield a
Variety mt kg/ha

El Paraiso 87 Traditional 91 128 767
39 Modern 65 68 855

Olancho 72 Traditional 93 1 1 7 71 3
54 Modern 54 102 441

a This estimate gives equal weight to each farmer’s mean yield (i.e. it is not the

division of the production over the area in this table).

Source: Farmer survey, 2002.

 

5.1.7. Bean Sales
Farmers grew beans for both home consumption and to sell. Therefore, farmers

can plant a specific variety for consumption only, for sales only or for both consumption

82

 

 

and to sell (Table 15). Results from the survey indicated that 100% of the farmers in El
Paraiso and 77% of farmers in Olancho sold part of their harvest.

The most common practice is for farmers to both consume and sell the variety that
he/she grows, as reported by 86% of the farmers in El Paraiso and 78% of farmers in
Olancho. Nevertheless, some farmers in both departments grew a specific variety (or
varieties) only for consumption (El Paraiso, 14%, Olancho, 22%) or only for sales (El

Paraiso, 17%, Olancho, 22%).

 

Table 15. Farmers (percentagea) preferred use of a variety (or varieties),
postrera 2000 to primera 2002, by department, Honduras.

 

 

 

 

 

Department
variety use El Paraiso (N = 76) Olancho (N = 75)
Consumption only 13.9 22.2
Sale only 16.7 22.2
Consumption and sale 86.1 77.8

 

 

 

aPercentages do not add to one hundred percent because farmers used
several varieties, each for different reasons, so a farmer can count twice
for one variety use.

Source: Farmer survey, 2002.

 

 

 

Farmers in El Paraiso sold almost twice as much of their 2001 harvest (57%), as
did farmers in Olancho (26%) (Table 16). This can be explained by Olancho farmers’
lower yields in 2001 (652.7Kg/ha, (1005 lb/mz)), compared to El Paraiso (860.4Kg/ha
(1048 lb/mz)). Thus, farmers in Olancho had less grain for sale, after setting aside grain

to meet their consumption needs.

83

Table 16. Farmers’ sales (percentage) of the 2001 postrera harvest, by type of variety
and department, Honduras.

 

 

 

 

 

 

Variety Department

El Paraisoa (N = 46) Olanchob (N = 41)
Traditional 75.2 (N = 32) 27.2 (N = 24)
Modern 36.3 (N = 14) 23.6 (N = 17)

 

a Total Harvest = traditional =34,484 kg, modem=28,996 kg
Total Harvest = traditional =29,727 kg, modem=l6,386 kg

Source: Farmer survey, 2002.

 

 

 

In Postrera, in El Paraiso 75% of the harvest of traditional varieties was sold,
compared to 36% of the modern varieties. However, this does not indicate that traditional
varieties are preferred for sale over modern varieties. Some key informants in Honduras
said that farmers plant both modern and traditional varieties, so they can consume their
preferred traditional variety and sell the modern variety, while others say that farmers sell
their traditional varieties because they command a higher price and consume the modern
varieties. However, the farmers surveyed showed no preference for selling either TVs or
MVs. In postrera of 2001, of the 28% of the farmers in El Paraiso who planted both types
of varieties, they sold 73% of their TVs harvest and 67% of their MVs harvest. In
Olancho, only 13.9% of the farmers planted both TVs and W5 in the same season and
they sold 15% of their TVs harvest and 25% of their MVs harvest.

5.1.8. Prices Received

Sales price data collected from the farmers (Table 17) confirms the conventional
wisdom that traders pay a higher price for grain of traditional varieties than for modern
varieties due to the former’s lighter red color, ranging fi'om 5.6% more in El Paraiso to

14.4% more in Olancho.

84

Table 17. Price (Lempiras/Kg) at which farmers sold their postrera 2001 harvest, by
type of variety and department, Honduras.

 

 

 

 

 

Variety Department
ElParaisoa(N=4l) OlanchoT’(N=21)

Traditional (N = 44) 8.67 8.12

Modern (N = 18) 8.21 7.10

 

 

 

’ Traditional N=29, bModem N=12
Source: Farmer survey, 2002.

 

 

Up to this point, several prices have been reported. Table 18 makes it summarizes

some grain and seed prices at various places that have been mentioned since Chapter 4.

Table 18. Summary of bean prices (in Lempiras /kg), Honduras, 2002.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Type of grain\ Place El Paraiso Olancho Tegucigalpa Zamorano
sold
Grain (2001 harvesta)
Traditional 8.67 8.12 NAb NA
Recycled modern 8.21 7.10 NA NA
Seed
Farmerc
Traditional 9.6 8.5 NA NA
Recycled modern 8.6 7.7 NA NA
Commercial
Hondugenet NA NA 33 NA
Certified
ZCE NA NA NA 35
DUWEST 38 38 36.5 NA

 

 

 

 

Not available

 

8 Grain sold following the postrera 2001 harvest.

c Price farmers expected to pay for seed bought from a neighbor.
Source: Farmer survey 2002 and key informant interviews.

 

 

85

5.2 Demand for Modern Varieties

5.2.1 Preferred Bag Size

In 2002, certified and commercial seed of modern bean varieties were sold in 50
lb bags (22.7kg bags) and 44 lb bags (20 kg bags). It was hypothesized that farmers
would prefer to purchase seed in a smaller bag. To assess the importance of bag size as a
constraint to purchasing seed, farmers were asked what bag size they preferred (i.e. 5 lbs,
10 lbs, 15 lbs, 20 lbs, 44 lbs, and 50 lbs (2.27 kg, 4.54 kg, 6.81 kg, 9.09 kg, 20 kg, and
22.7 kg, respectively)). While most farmers (57%) preferred a 50 lb bag, 26% preferred a
20 lb bag. In spite of the fact that farmers did not report bag size as a reason not to
purchase seed, once it is presented as an option, some farmers stated the preference of a
20 lb bag (26%). This low percentage of farmers stating a preference for a 20 lb bag but
who also would buy a 50 lb bag (3 5%) suggests that seed should be sold in 50 lb bags.

5.2.2 Purchase Frequency

To estimate the annual demand for improved seed and also to obtain an estimate
of how often farmers like to renew their seed stock, farmers were asked how long after
they have purchased improved seed would they buy again. The average time between
purchases of seed was of 1.7 years for those who were willing to buy.

5.2.3 Price Discounts for Modern Varieties

To assess the market acceptability of the both existing and yet-to—be-released
modern varieties, traders in Tegucigalpa were asked what price they would pay for each
of six MVs. The six modern varieties were Catrachita, Dorado, Tio Canela 75, Milenio

(not yet released in July 2002), Amadeus 77 (not yet released in July 2002), and

86

Carrizalito (not yet released in July 2002). Catrachita, a MV with market qualities similar
to TVs was used as a control.
The prices that traders in “Mercado Las Americas” estimated that they would pay

for these six varieties are listed in Table 19.

Table 19. Traders (N = 5) procurement prices of modern varieties in Mercado Las
Americas, July 2002, Tegucigalpa, Honduras.

 

 

 

 

 

 

 

 

 

 

 

 

Variety Price (L./kg) Price Index Coldr
Catrachita (control) 9.0 100 Light red
Dorado 8.2 91 Reddish black
Tio Canela 75 8.5 94 Dark red
Milenio 9.4 104 Light red
Amadeus 77 9.4 104 Red
Carrizalito 8.3 92 Light dark red
Source: Key informant, 2002.

 

 

As expected, traders were willing to pay a higher price for lighter red color grain.
However, in addition traders take into account freshness and cleanliness of the grain.
Traders were willing to pay higher prices for Milenio and Amadeus 77. They have high
prices due to their lighter color-compared to Dorado and Carrizalito. Since Amadeus
was a soon-to-be-released variety, these data suggest that this variety will be highly
acceptable to farmers.

5.2.4 Farmers Preferences for Recent Modern Varieties

To assess farmers’ preferences for the six recently-released varieties, the farmers
were shown an unlabeled sample of the varieties and asked to rank each variety. The

farmers’ rankings were used to create a global ranking score for each variety (Table 20).

87

This was done by assigning a value of six to the variety that farmers ranked highest, five

for the second highest ranked variety, etc. The 72 farmers rankings of each variety were

summed to obtain an overall ranking score for each variety.

Table 20. Ranking of six recently released varieties, by department, July 2002,

 

 

 

 

 

 

 

 

 

 

 

 

 

Honduras.
Place in ranking El Paraiso Olancho

Variety Score Variety Score
First Amadeus 77 172 Amadeus 77 184
Second Catrachita 140 Catrachita 144
Third Milenio 123 Carrizalito 13 1
Fourth Carrizalito 118 Tio Canela 75 119
Fifth Dorado 108 Milenio 109
Sixth Tio Canela 75 95 Dorado 69
Source: Farmer Survey, 2002.

 

 

In both departments, the farmers assigned the highest ranking to Amadeus 77 and

Catrachita and the lowest ranking to Tio Canela and Dorado in El Paraiso and Olancho,

respectively.

As noted earlier, many farmers prefer TVs because they command a higher market

price. The results in the table above indicate that the farmers assess Amadeus 77's quality

characteristics (i.e., color) as superior to all other modern varieties—including Catrachita,

which is similar in color to TVs.

In addition, farmers were asked to state the price that they would be willing to pay

for seed of each variety. These data were used to estimate the percent of farmers who

would be willing to pay a price ranging from L 9/lb (L l9.8/kg)(minimum price thought

88

 

that a bean seed farmer would be willing to sell) to L 17/lb (L 37.4/kg)(expected certified
seed price) for each of the six varieties, by department (Figure 8 and 9). For each variety,
each positive response of a farmer to a price for an specific variety was added to the
percentage of farmers willingness to pay. For prices lower than L 17/lb, the reported
percentage includes the accumulated positive responses up to that price, since if someone
is willing to pay L 17/1b they would also be willing to pay a lower price.

In both departments, Amadeus 77 is the variety with the highest demand. Figure 8
and 9 show that 70% and 81% of the farmers in El Paraiso and Olancho, respectively, are
willing to pay L 9/lb of seed of the variety Amadeus 77. In El Paraiso, Dorado follows
Amadeus 77 with the second highest demand where 39% of farmers are willing to pay L
9/lb of seed. On the other hand, in Olancho, Carrizalito and Tio Canela 75 follow
Amadeus 77 with the second highest demand where also 39% of farmers are willing to
pay L 9/lb of seed for each variety. These shows the farmers’ preference for Amadeus 77.

Farmers’ preferences for Amadeus 77 makes it a variety with high potential
demand. DUWEST and small-scale farmers participating in the seed scheme would be
more certain that its seed would sell. This high demand can be attributed to a lower than

usual price for commercial seed (L 15/lb), and to the light red color of the seed.

89

Figure 8. Farmers’ willingness to pay for bean varieties in El Paraiso, 2002, Honduras.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

18
17 '\ —-— Amadeus 77
16 -
15 . \’ k —«— Carrizalito
14 .
L.llb 13 \ + Milan”

12
11 \\\L\\\\ \\ + Dorado
10 r + T‘io Canela 75

9 ~M \-

8 . I 1 . + Catrachita

0 20 40 60 80
% of farmers

 

 

 

Figure 9. Farmers’ willingness to pay for bean varieties in Olancho, 2002, Honduras.

 

18-

17 _ \ + Amadeus 77
16 C . .
46.x -- m
14 -—>.—— Milenio
L./lb 13 “M \1

1 2 \\ + Dorado

n 1( .1 Q
1 0 ., + Tio Canela

9 M \u 75

+ Catrachita

 

 

 

 

 

 

 

 

 

 

 

0 20 40 60 80 100

% of farmers

 

 

 

90

To estimate an annual demand curve for each variety (Figures 10 and 11), farmers
responses regarding their preferred bag size and frequence of bean seed purchase were
combined with their willingness to pay. So for example, if a farmer responded that he/she
was willing to pay L lS/lb, his/her preferred bag size was 44 lbs, and his/her frequency of
bean seed purchase is once every two years, then his/her demand is of 22 lbs of seed per
year at a price of L 15/lb (441b*1/2 purchases/year). Analogous to the demand
percentage, for prices lower than L l7/1b, the reported demand includes the accumulated
demand up to that price.

While farmers in both departments preferred Amadeus 77 above all other
varieties, the demand curve is more elastic in El Paraiso than in Olancho. Since farmers
would likely only purchase seed of a single new variety, these curves represent an “either
or” demand for the varieties (e. g. farmers would not demand all of them at the same
time). Results from the farmer ranking of the varieties and the willingness to pay analysis
are generally consistent, except for Catrachita. While farmers ranked Catrachita as the
second most preferred variety, the demand for Catrachita was low and inelastic.

However, willingness to pay (rather than rank) better reflects farmers preferences because
it forces farmers to assign a price to each variety.

Assuming a seed rate of 32 kg/ha, the sampled farmers in El Paraiso and Olancho

annually require approximately 4,02023 kg (8,844 lbs) and 3,203“ kg (7,046 lbs) of seed,

 

23 (2.12+1.37)*36*32 = 4020.48 where (2.12+1.37) is the average area planted by each
farmer (Table 8), 36 is the number of farmers in the department and 32 kg/ha is the seed
rate.

2" (2.02+0.76)*36*32 = 3,202.56 where (2.2+0.76) is the average area planted by each
farmer (Table 8), 36 is the number of farmers in the department and 32 kg/ha is the seed

9]

respectively. These figures are calculated by multiplying the sampled farmers average
bean planting area per year (sum of primera and postrera areas (Table 8)) by the assumed
seeding rate of 32 kg/ha times the number of farmers in the department (i.e. 36). So the
total seed requirements for the sampled bean farmers in El Paraiso the estimation and
estimate would be the following: (2.12 ha in postrera+l .37 ha in primera)*36 farmers“32
kg/ha (seeding rate) = 4020.48 kg of seed.

Figure 10. Sample farmers annual demand by variety in El Paraiso, 2002, Honduras.

"\

15-

LJlb 13 g
\t

11

 

+ Amadeus 77

—-— Carrizalito

 
  

 

-—>.a—— Milenio

 

+ Dorado

\ +10 nea
\ 15°“
\

+Catrachita
\ \\. \\n .

0 100 200 300 400 500 600 700 800 900
lbs of seed

 

 

 

 

 

 

 

 

 

 

 

 

 

 

9

 

 

 

 

 

 

rate.

92

Figure 11. Sample farmers annual demand by variety in Olancho, 2002, Honduras.

17 a
\ —I—- Amadeus 77
\

1 5 \_ \_ -—-— Carrizalito

N\T\\ \\ —><— Milenio

\.
L.Ilb 13 \ \
l l + Dorado
\ + Tlo Canela 75

 

 

 

 

l
11 i

I
9 I ll . -0..-

v - I

0 100 200 300 400 500 600 700

 

_,.«

 

 

 

 

 

 

 

 

 

 

 

 

lbs of seed

 

 

Table 21 shows the estimated amount of seed that the sampled farmers were
willing to purchase at various prices and the departments demand. Farmer demand per ha
is estimated by dividing the seed demand at each price obtained from Figures 10 and 11
by the number of farmers in each department (i.e. 36). For example in El Paraiso, the
total demand of Amadeus seed of the sampled farmers at L 17/lb is of 48 lbs. This
estimate of 48 lbs is then divided by 36 (number of farmers sampled in El Paraiso) to
obtain 1.33 lbs/farmer. This latter estimate, average individual farmer demand, is then
divided by the sample farmer average annual bean area obtained by a summation from
Table 8, 3.49 (ha in El Paraiso (2.12 ha in postrera + 1.37 ha in primera)) to consequently
obtain demand per ha. Finally, demand for the department is estimated by multiplying

demand/ha times the number of hectares planted in postrera 2001 and primera 2002 in

93

each department (seasons where average area was calculated from) as reported in the

National Statistics Institute of Honduras (INE).

Table 21. Estimated annual demand (kg) for seed of Amadeus 77 at various prices, by

 

 

 

 

 

 

 

 

 

 

 

 

 

department, Honduras, 2002.
Department El Paraiso‘1 Olanchob Total
----_- ----- (kg)
Pme Ptr Sample Demand in Sample Demand in
pound (kg/farmer/ha) department (kg/farmer/ha) department
17 0.38 7,566 1.08 23,518 31,083
15 0.68 13,595 1.54 33,474 47,069
13 0.83 16,553 2.47 53,777 70,331
11 1.57 31,173 2.51 54,718 85,891
9 2.90 57,510 2.99 65,066 122,576
a Average area planted in El Paraiso is 3.49 ha. Farmer survey, 2002.

Average area planted in Olancho is 2.78 ha. Farmer survey, 2002.
Source: Farmer survey, 2002.

 

 

 

According to the most recent Agricultural Census (1993), farmers in El Paraiso
and Olancho planted an average of 1.24 ha and 0.97 ha of beans, respectively. The fact
that the sampled farmers in El Paraiso and Olancho planted and average of 3.49 ha and
2.78 ha to beans, respectively, suggests that these farmers were likely more interested in
bean production, compared to the average bean farmer. Thus, the estimate of department
demand is likely biased upwards. However, since demand is calculated based on 2002
actual hectares, the bias is likely to be capped. Also, Mather’s (2003) farmer average total
bean area (primera and postrera) for Mideast (El Paraiso and Francisco
Morazan)/Northeast (Olancho) adds to 3.65 ha, which is similar to the results in this

research sample, and thus the bias is thought to be minimal.

94

Consequently, the total potential annual demand for Amadeus 77 in El Paraiso,
ranges from 7.6 mt (if sold at L 37/kg (L 17/lb)) to 57.5 mt (if sold at L 20/kg (L 9/1b)).
In Olancho, ranges from 24 mt (if sold at L 37/kg (L 17/lb)) to 65 mt (if sold at L 20/kg
(L 9/lb)). Total demand in both departments ranges from 31 mt (if sold at L 37/kg (L
17/lb)) to 123 mt (if sold at L 20/kg (L 9/lb)).

5.2.5 Factors Associated with Farmer Demand

A double hurdle model was used to assess variables associated with farmers desire
to purchase each variety and their WTP. The explanatory variables in the model and the
expected sign for each variable are presented in Table 22. Both varietal traits and farmer
characteristics are hypothesized to affect farmers desire to purchase and WTP. Regarding
varietal traits, the model could have included grain size, weight, and color because those
are characteristics that the farmer can evaluate by observing the grain samples.
Nevertheless, size and weight were not included because of the similarity of the varieties
and the difficulty to distinguish them without an instrument. Grain color (lighter red
colors) is expected to increase the probability to purchase and WTP because middlemen

pay a higher price for light-red grain.

While characteristics like cooking time and flavor often influence farmers’
varietal choices, they were not included because they could not be observed. Also,
production characteristics (e. g. yield, disease resistance) are not included in the model
since farmers could not observe these traits. However, farmers were told to assume that

the varieties were disease resistant and would yield 1300 kg/ha, like Tio Canela.

95

Regarding socioeconomic characteristics, age was included as a proxy for farming
experience; education to measure capability to understand the advantages of different
varieties; land to measure access to capital (i.e. wealth); prior use of modern varieties to
measure experience planting and awareness of modern varieties, and fertilizer use to
measure adoption of recommended management practices. However, fertilizer and
modern variety use are predetermined variables, since fertilizer use is for a previous
season (Postrera 2001) and modern variety use indicates if the farmer had used a modern

variety at least once in any of the past four seasons (i.e., Postrera 2000 to Primera 2002).

Table 22. List of explanatory variables in determining desirability to buy (DB) and
willingness to pay (WTP) and the hypothesized relationship between these explanatory
variables and farmers’ decision to buy and WTP.

 

 

 

Variable Unit of measure Expected change in
purchase decision
and WTP

Variety attributes
Color Color (reddish black, dark +

red, light dark red, red, light

red) (goes from very dark to

light red)a.

 

Farmer’s management and
Socioeconomic characteristics

 

 

 

 

 

Age Years -
Education Years in school +
Land ha +
Fertilizer used in Postrerab 2001 0 = no 1 = yes +
Use of modern varieties 0 = no 1 = yes +

 

 

 

a Change from black through light red is suppose to increase DB & WTP.

b Postrera is the rainy season from September to November.

 

 

 

96

Farmer’s age is expected to reduce the probability to purchase since it is widely
assumed that older farmers stay loyal to their cmrent varieties. Education is expected to
increase the probability to purchase and WTP because it is associated with capability of
being able to differentiate between varieties. Land, fertilizer use in previous seasons, and
use of modern varieties in the past are expected to increase the probability to purchase
and WTP because these variables are associated with commercial orientation (compared
to subsistence farming), better management practices, and awareness of benefits of

modern varieties, respectively.
5.2.6 Results

The final model” included color, age, education, land, and fertilizer used in
postrera 2001 as explanatory variables. The probit model correctly predicted 72.45% of
the outcomes (Appendix B). The marginal effects or change in probability, due to a
change in one unit in the explanatory variables for the probit and truncated regressions,

are shown in Table 23. The probit and truncated result tables are in Appendix C.

 

2’ Correlation matrix is in Appendix A.

97

Table 23. Marginal effects of bean attributes and farmer’s characteristics on purchase
decision and willingness to pay (WTP), Honduras, 2002.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variable Probit Model Truncated Regression
Dark red color -0.002 0.342
Std. error 0.084 0.909
p—value 0.975 0.706
Lighter dark red color 0.106 —0.178
Std. error 0.085 0.812
p-value 0.214 0.826
Red color 0.488' ' ° a 0.403
Std. error 0.072 0.760
p-value 0.000 0.596
Light red color -0.008 0.033
Std. error 0.072 0.822
p-value 0.917 0.968
Fertilizer used in Postrera 2001 -0.056 -0.677
Std. error 0.057 0.544
p-value 0.322 0.213
Age 0.002 0.053' '
Std. error 0.002 0.023
p—value 0.400 0.02]
Education 0.005 0.224' ' '
Std. error 0.007 0.061
p-value 0.472 0.000
Land -0.003 0.094' ° °
Std. error 0.004 0.023
value 0.3 73 0.000
Use of modern variety 0.051 1.487'"
Std. error 0.053 0.578
p-value 0.334 0.010
N= 432 , 72.45% of predicted outcomes are correctly predicted in the probit model.
a ° Significant at a 10% level, " Significant at a 5% level, and ... Significant at a 1% level.

 

 

First, a probit regression was carried out to observe the effect of color of the grain

and farmers’ characteristics on the farmers’ purchase decision. In the probit regression,

98

red color was the only significant variable. The change in color from a reddish black to a
red increases the probability to purchase a modern variety by almost one-half point (i.e.
50%). This is a high increase in probability. This color is associated to the color of
Amadeus 77 while the reddish black color is associated with Dorado, the most widely
adopted variety. This indicates that Amadeus appeals to farmers due to its color.
Nevertheless, a lighter red color (e.g. Milenio and Catrachita) was not significant and had
a negative sign, contrary to expectation. This could be due to the fact that Milenio and
Catrachita have a pale red color. Finally, Carrizalito, which had a darker color than
Amadeus 77, should not be released because the results indicated that its color did not

impacted farmers’ purchase decision.

The double hurdle model exemplifies the importance of separating the purchase
decision from willingness to pay, since color was a major determinant in the purchase
decision, while in willingness to pay socioeconomic characteristics were more important

than varietal characteristics.
5.3 Summary

A farmer survey was conducted in El Paraiso and Olancho to collect
socioeconomic characteristics of farmers and to assess the demand for improved bean
seed. Also, farmers’ seed use, production, and sales information was obtained. From
farmers’ socioeconomic characteristics, farmers’ land ownership, age, and years in school
was collected. Regarding seed use, production, and sales, information on bean area,
source of bean seed, varieties planted, total production, bean sales, and prices was

collected. Finally, a double-hurdle model was used to assess demand for modem

99

varieties as well as some demand curves that were constructed using information on

preferred bag size, purchase frequency, and preferences on recent modern varieties.

The sampled farmers owned an average of 5.37 ha of land, had an average of 51
years of age, and had completed the fourth grade. Regarding the bean area planted,
farmers in El Paraiso and Olancho grow an average area to beans in the primera season of
1.37 and 0.76 ha, respectively. In the postrera season, farmers in El Paraiso and Olancho

grow an average area of 2.12 and 2.02 ha, respectively.

Farmers obtained bean seed from several different sources and planted 22 bean
varieties. Farmers in El Paraiso and Olancho obtained bean seed mainly from saved seed
from previous harvest (73% and 63%, respectively) and from another farmer (20% and
26%, respectively), and only a few farmers obtained it from NGO/Projects, certified seed
dealer, or purchased in grain market. The most quoted reason (61% of farmers) for not
buying certified seed was that it is too expensive. Also, farmers in El Paraiso and
Olancho planted 10 and 18 bean varieties, respectively, for a combined total of 22 bean

varieties.

Farmers in El Paraiso and Olancho obtained yields of 767 kg/ha and 713 kg/ha
with traditional varieties, respectively, and 855 kg/ha and 44] kg/ha with modern
varieties, respectively. From the harvest obtained, farmers in El Paraiso and Olancho
commonly both consume and sell the grain (86% and 78%, respectively). Nevertheless,
some farmers grow specific varieties only for consumption or only for sales. Regarding
bean sales, farmers in El Paraiso sold twice as much of their harvest (5 7%) compared to

farmers in Olancho (26%). Also as expected, in El Paraiso and Olancho, modern

100

varieties were sold at a lower price (L 8.67/kg and L 8.12/kg, respectively) compared to

the traditional varieties (L 8.21 and L 7.10/kg, respectively).

Finally, regarding the demand for modern varieties, most farmers (57%) prefer to
buy bean seed of the varieties shown to them in 50 lb bags. Also, farmers average
renewal time of varieties shown to them was of 1.7 years. From the six varieties shown
to farmers, Amadeus 77 was the highest ranked by farmers and with the highest demand
in both departments. The double-hurdle model used to assess demand reveals the
importance of red color in farmers’ purchasing decision. Socioeconomic characteristics

like “land ownership , education”, and “previous use of modern varieties” were

associated with farmers willingness to pay.

101

Analysis presented previously indicates that farmers in El Paraiso and Olancho are
interested in purchasing seed of the modern variety Amadeus 77. Drawing on
information reported regarding options for producing and marketing improved bean seed,
this chapter proposes a strategy for implementing the proposed bean seed multiplication
and distribution scheme, in which small-scale farmers will produce the seed and sell it to
DUWEST, as summarized in Figure 12. However, as will be discussed later, DUWEST

is only willing to buy 23 mt of seed in year one. Thus, the target level of production is

Chapter VI.

much lower than the estimated demand.

6. Bean Seed Production Scheme

Figure 12 . Proposed bean seed production scheme.

Breeding

 

 

PIF

 

 

 

 

 

 

 

ARSAGRO farmers

 

 

 

CERTISEM

 

 

Multiplication

 

 

 

Yaruca farmers

 

 

 

Inspection

 

 

 

 

 

 

 

 

DUWEST
Subdistributors

 

 

Marketing

102

 

 

 

DUWEST

 

 

 

6.1 Supervision of the Bean Seed Scheme

Dr Juan Carlos Rosas (PIF/Zamorano), who has expressed an interest in providing
leadership for implementing the scheme, will supervise the bean seed scheme. That is, he
will be responsible for training of the farmers, supervision of the farmer fields, and final
approval of the seed to be sold to DUWEST. Zamorano will provide a certificate of good
quality for the farmer groups’ seed, so that DUWEST can trust the quality of the seed

from these farmer groups.
6.2 Seed Production Sites and Participants

Two farmer groups, located in El Paraiso (ARSAGRO) and Atlantida (Yaruca
farmer group), will be trained to produce commercial seed for sale to DUWEST.
ARSAGRO farmers in El Paraiso are proposed due to several reasons. ARSAGRO
farmers have experience growing seed, have access to credit and machinery by being a
member of the association, and are located close to Zamorano. The Yaruca farmer group
is proposed due to different reasons. Yaruca farmers can grow seed in the Apante season,
which gives them the advantage of being able to produce seed closer to the planting
seasons. Also, Yaruca farmers are supported by the Atlantic Litoral Region University
Center (CURLA), which in turn collaborates with PIF. Therefore, extension services and

supervision of farmer plots will be made easier.
6.3 Contracts

Two types contracts will be required to establish each participant’s responsibility--
one between the farmer groups and PIF, and another between the farmer groups and

DUWEST.

103

The contract between farmer groups and PIF will establish PIF’s responsibility to
train and supervise the farmers through at least 7 visits to the farmers fields. PIF will also
extend a certificate26 of good quality, if farmers follow PIF’s recommendations and the
seed meets the required quality standards. The farmers will promise to pay PIF a fee for
the training and supervision of the plots affer they are paid by DUWEST (120 days afier

planting).

The contract between farmer groups and DUWEST will commit DUWEST to
purchase 17 mt of commercial seed from ARSAGRO farmers and 6 mt from the Yaruca
farmer group at a guaranteed pre-planting price to be paid 30 days after delivery,
conditional on it being certified by PIF and CERTISEM. The farmer groups will promise
to deliver the specified amount of commercial seed in sealed bags, at the department
capital, no later than 100 days after planting. This delivery date insures enough time to
process and bag the seed after CERTISEM laboratory test results are available. To insure
that the farmers sell the seed to DUWEST, DUWEST will pay the plot registration fee
and for CERTISEM field visit. This cost will be deducted from the payment that
DUWEST makes to the farmers. The contract will also specify that DUWEST is the

exclusive seed distributor for these farmer groups.
6.4 Credit

ARSAGRO farmers will obtain credit from their association, which extends credit

to its member for up to 24 months at a monthly interest rate of 5%. While this rate is

 

2" Zamorano’s management has no objection for the seed to be labeled “Zamorano
Approved Seed”, since the label would only guarantee that it meets physical and variety
characteristics.

104

higher than the interest rate of other commercial banks, it has the advantage of requiring
less paperwork and time is shortened. On the other hand, the Yaruca farmer group will
obtain credit from a bank to pay for the registered seed and inputs« either BANADESA

or Banco de Occidente due to their orientation to work with small-scale farmers.
6.5 Commercial Seed/Variety
Both farmer groups will purchase registered seed27 from PIF/Zamorano. Amadeus

77 will be the variety produced and distributed under the scheme because of its high
productivity and high ranking of the varieties by El Paraiso and Olancho farmers. PIF is
willing to sell up to 9,000 kg of registered seed at a price of L 2,000 per 45 kg bag.
However, since DUWEST will only buy 23,000 kg of commercial seed and expecting
approximately an average yield of 1100 kg/ha, only a total of 21 ha will be planted.
ARSAGRO farmers will plant 15 ha and Yaruca farmers will plant the other 6 ha.
ARSAGRO farmers will be allocated a larger quota than the Yaruca farmers because of

their greater seed production experience.
6.6 Farmer Training and Supervision

To insure high seed quality, Zamorano/PIF will train and supervise the farmer
groups in bean seed production. Training will only take place during the first season.
After the first season, PIF will only do supervision visits. The farmers will be trained in

the seed producing techniques, especially disease control and threshold pest levels in

 

27 Regarding the use of the variety Amadeus 77 (property of Zamorano), Zamorano will
not collect intellectual property rights royalties because Zamorano is trying to promote
widespread distribution and adoption of its modern varieties.

105

order to know the right time to spray. Training and supervision visits will be scheduled
for the following periods: land plowing, planting, fertilization, flowering, grain filling,

harvest, and selection and bagging.

In the first season, both training and supervision will be done on the same visits to
the farmers. PIF will charge $30/training day, plus transport and lodging costs—which are
discounted rates considering that Zamorano normally charges $100/person/day plus
transport and lodging costs, and that the Honduran Agricultural Research Foundation

(F HIA) charges $80/person/day.

In subsequent seasons, PIF will charge $30 per day, plus transport and lodging

costs for routine inspections during the bean growing season.
6.7 Seed Processing/Packaging

The ARSAGRO farmers will be responsible for seed drying, cleaning and
selection in the fields. Then, the seed will be taken to ARSAGRO’s office in Danli,

where it will be inspected by CERTISEM. After CERTISEM provides the seed labels,

fungicide and insecticide will be applied to the seed prior to bagging.

The Yaruca farmers will also be responsible for drying, cleaning and selection in
the fields. Then they will take the seed to their village seed bank to await seed inspection
by CERTISEM. Similarly, after CERTISEM provides the seed labels, fungicide and

insecticide will be applied to the seed prior to bagging.

106

The bean seed will be marketed in 50 lb bags, due to farmers response that this
was their preferred size bag. Although the bag will not carry Zamorano University logo,

it will have a “Zamorano Approved” sticker.
6.8 Seed Inspection

CERTISEM will carry out the final seed inspection and provide the seed label
certifying that it meets commercial seed requirements. This will be done by CERTISEM
staff taking random samples of the seed and rtmning germination and purity tests. After
approved by CERTICEM, PIF will issue the certificate of good quality seed to the

farmers.
6.9 Distribution

After the seed is bagged and labeled, the farmer groups will transport the seed to
subdistributors in the closest department capital-Danli for ARSAGRO farmers and La
Ceiba for the Yaruca farmers. DUWEST will likely later leave a portion of the seed in
those cities and also distribute some of it to J uticalpa and San Francisco de la Paz in
Olancho and to the city of Yoro and the town Yorito in the department of Yoro. Also, the
seed obtained from the Yaruca Farmer Group could be distributed to other bean

producing regions such as Colon.
6.9.1 Price

DUWEST will pay the farmer groups L 8/lb for the seed. Assuming a yield of
1,090 kg/ha, ARSAGRO farmers can produce seed for L 4.60/lb (L 10.12/kg) and the

Yaruca farmers can produce seed for L 4.95/lb (L 10.89/kg), excluding training or

107

supervision costs (Table 24). Including training cost for the first year and supervision
costs for the second year, in the first year the cost per pound for the ARSAGRO and
Yaruca farmers is L 4.87/lb (L 10.71/kg) and L 7.14/1b (L 15.71/kg), respectively, and in
the second year the cost is L 4.8l/lb (L 10.58/kg) and L 6.59/lb (L 14.50/kg), respectively
(Table 25). Assuming that ARSAGRO and Yaruca farmers plant 15 ha and 6 ha,
respectively, DUWEST’s purchase price of L 8/lb is very appealing for both farmer
groups. ARSAGRO farmers and Yaruca farmers would need to produce 1,461 lbs/ha
(664 kg/ha) and 2,142 lbs/ha (974 kg/ha), respectively, to break even in the first year, and
would need to produce 1,443 lbs/ha (656 kg/ha) and 1,977 lbs/ha (899 kg/ha),

respectively, to break-even in the second year.

DUWEST will markup the purchase price by 20-25% (which includes
transportation costs to the towns (3%) and its subdistributor in each town will markup the
seed by an additional 10-15%. Thus the retail price at small towns like Juticalpa, Salama,
and San Francisco de la Paz in Olancho and Danli, Linaca, and Chirinos will be of

approximately between L “Ab and 12/1b (Table 26).

At a price of L 12/lb, the estimated seed demand in El Paraiso and Olancho is of
24 mt and 54 mt, respectively (Table 21). However, DUWEST is only willing to
purchase 23,000 kg—which is equal to approximately 30% of the combined total demand
in both departments. This suggests that DUWEST will face minimal risk of not being

able to sell its stock of seed.

108

Table 24. Estimated commercial seed production costs in El Paraiso and Atlantida, Honduras, 2002.

Activity
Land Preparation
Plow
Disk
Cultivator
Thresher
Inputs
Seed
Fertilizer (12-24-12)
Foliar
Herbicides
Gramoxone
Fungicides
Poliran
Captan
Insecticides
Decis
F olidol
Labor
Manual land prep.
Planting
Fertilizing
F oliar
Herbicide
Insecticide
Weeding
Roguing
Harvest
Beating
Transport field-home
Processing
Drying
Cleaning
Selection
Bagging
Bag cost
Tag
CERTISEM
Transport
House- Dept. capital
Opportunity Cost
Land

unit

rental
rental
rental
rentang

kg
k8
It

It

k8
Re

It
It

man-day
man-day
man-day
man-day
man-day
man-day
man-day
man-day
man-day
man-day

kg

man-day
man-day
man-day
man-day
units

kg

ha

km

ha

Interests for 4 month period %

TOTAL

n.a.
n.a.
n.a.

n.a.
l 1
29
14

1,182

NQW-fi

48

20

l

2,483c

ARSAGRO
units/ha cost/unit cost/ha

643
360
630
0.55

44

120

n.a.

90
60

150
n.a.

n.a.
n.a.
n.a.
40
n.a.
40
40
40
10
40
0.1 l

40
40
40
40

20
104

35

l 000

0.20

643
360
630
600

1,408
539
240

n.a.

180
60

300
n.a.

n.a.
n.a.
n.a.
240
n.a.
440
1,160
240
140
120
130

160
120
280
80
180
480
104

700

l ,000

497

11,031"

a EMS is not applicable; b Assumes a yield of 2,400 lbs.;
c Cash expenses (excludes land preparation, labor, and seed); 6 cost is L 4.60/lb;

fcost is L 4.95/lb;

Source: Key informants 2002.

109

a
[1.3.

n.a.
n.a.
n.a.

32
136
2

14
10

1,182

30

1

3,616

Yaruca
units/ha cost/unit cost/ha

n.a.
11.8.
11.8.
n.a.

44
4
120

100

90
60

n.a.
400

50
50
50
50
50
50
50
50
50
50
0.22

50
50
50
50
4
20
137

50

450
0.07

n.a.
n.a.
n.a.
n.a.

l ,408
539
240

100

l 80
60

n.a.
200

850
500
300
100
150
200
1,500
300
700
500
260

200
150
350
100
180
480
137

1,500
450

253
11,887

1'

Table 25. Seed production training and inspection costs of farmers fields, Honduras,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2002.
Activity Farmer Group Days Cost/day Total
(5)
Training ARSAGRO
Year 1 Fee 8 30 240
Meals in Danli" 8 15 120
Transport Zamorano-field 8 30 240
Total 600b
Yaruca
Fee 8 30 240
Lodging and meals6 16 55 880
Transport 8 100 800
Total r920d
Inspections ARSAGRO
Year 2 Fee 6 30 180
Lodging and mealsc 6 15 90
Transport 6 30 180
Total 450f
Yaruca
Fee 6 30 180
Lodging and meals8 12 55 660
Transport 6 100 600
Total 1440h

 

 

 

 

 

 

a For one person; b Cost is $600*L16.44/$/(15ha * 2,4001bs/ha) = L0.27/lb;

c For one person, one night; d Cost is $1920*L16.44/$/(6ha * 2,4001bs/ha) = L 2.19%

e For one person; fCost is $450*L16.44/$/(15ha * 2,4001bs/ha) = L 0.21/lb

8 For one person, one night; h Cost is $1440*L16.44/$/(6ha "‘ 2,4001bs/ha) = L 1.64/lb
Source: PIF, 2002.

 

110

 

Table 26. Estimated prices for commercial seed by stage in proposed bean seed
scheme, Honduras, 2002.

 

 

 

 

 

 

 

Stage Cost Total Price
(Lempiras/lb) (Lempiras/lb)

Price DUWEST will pay farmers for commercial 8 8

(inspected and bagged) seed at department capital

DUWEST’s markup (22%) on seed purchased from 1.76 9.76

farmers at department capital

Transport cost from department capital to smaller 0.29 10.05

town or to department capital in another departmenta

(3%)

DUWEST subdistributor’s markup (15%) on seed 1.51 11.56

supplied by DUWEST to smaller towns

 

 

a Smaller towns refer to towns like Chirinos and Linaca; department capitals refer to
towns like Juticalpa and Yoro.
Source: Key informants, 2002.

 

111

 

Chapter VII
7. Summary, Conclusions, and Policy Implications

7.1 Summary

In Honduras, new varieties have been released with high yield potential and
resistance to various diseases, especially Bean Golden Yellow Mosaic Virus (BGYMV).
On-farm trials in Honduras indicated that recently-released modern varieties (MVs) like
Tio Canela produce higher yields than traditional varieties (TVs). A recent study found
that over 50% of bean farmers in the main bean-producing areas of Honduras (El Paraiso,
Olancho and Francisco Morazan) planted MVs, which indicates that they see advantages
to planting MVs-primarily due to their higher yield potential and disease resistance.
However, many of the farmers planted varieties that were released in 1990 and virtually

all of the farmers obtained their seed from a neighbor or planted farmer-saved seed.

The general objective of the study was to assess the feasibility of both increasing
the physical supply of improved seed of modern bean varieties and expanding farmer
access to these varieties. In order to recommend actions to increase the productivity of a
seed system, it is necessary to take into account the varietal characteristics preferred by
farmers and their willingness to pay and to evaluate alternative bean seed production and
marketing schemes that can deliver good quality seed at low prices. A seed system model
is proposed in which small farmers are contracted to multiply seed of modern bean
varieties, under the supervision of Zamorano, and the seed is marketed through a private

sector firm that currently distributes agricultural inputs.

112

A rapid appraisal methodology was used to collect information from key
informants required to describe the bean seed subsector in Honduras, identify potential
bean seed producers, and identify marketing agents interested in selling the seed. In
addition, a farmer survey was conducted to assess the potential demand for improved

bean seed.

The bean seed supply in Honduras is characterized by a series of seed production
stages leading up to the distribution of seed through a marketing network. The first stage
is breeding, followed by seed multiplication, seed inspection, and finally sales and
distribution of seed. The breeding is done by Zamorano/PIF and DICTA. Following this,
seed multiplication is done by Zamorano (PIF, ZCE, and ZCE’s contractees), DICTA,
Hondugenet, and NGOs (including NGO participating farmers). Seed inspection is done
by CERTISEM. Finally, seed distribution and sales are done by Zamorano/ZCE,

Hondugenet, and NGOs.

Two farmer groups, located in El Paraiso (ARSAGRO) and Atlantida (Yaruca
farmer group), expressed an interest in growing bean seed under contract. A bean seed
budget was constructed to estimate the cost of producing commercial seed. This analysis
found that commercial bean seed could be produced for L10.58/kg (ARSAGRO farmers
cost in second year) and L14.50/kg (Yaruca farmer group cost in second year), and
compared to the price of L 17.6/kg that farmers would sell their seed, they considered it

very attractive.

113

A farmer survey was conducted in El Paraiso and Olancho to collect

socioeconomic characteristics of farmers and to assess the demand for improved bean
seed, including their assessment of six modern varieties and their willingness to pay for
these varieties. To assess farmer demand for modern bean varieties, a double hurdle
model was used to determine the degree and direction of influence that farmers’ and
varietal characteristics have on farmers’ purchase decision and willingness to pay. In
addition, farmers’ preferred bag size, purchase frequency and willingness to pay was used

to appraise bean seed demand.

Among the six varieties shown to the farmers, farmers in both departments ranked
Amadeus 77 highest in terms of preferences. The double-hurdle model confirmed the
importance of red color in farmers’ purchasing decision and that socioeconomic
characteristics (i.e. land ownership, education, previous use of modern varieties) were
associated with greater farmer willingness to pay a premium price for a modern bean
variety. The farmers preferred a 50 lb bag and on average reported that they would

purchase new seed every 1.7 years.

While several input distribution firms were contacted, DUWEST was the only
marketing firm interested in participating in the proposed bean seed scheme. DUWEST
agreed to initially purchase 23,000 kg of improved seed and distribute it to its sub-

distributors.

Finally, this study demonstrated that it is feasible to contract small-scale farmers
to produce commercial seed of modern varieties and sell the seed through an input dealer

at department capitals at lower prices (L 12/lb (L 26.4/kg)) than the normal selling price

114

of L1 5/lb (L 33/kg) for commercial seed sold in Tegucigalpa, L 15.90/lb (L 35/kg) for
certified seed in Zamorano, or from L 17.27 (L 38/kg) to L 19.86/1b (L 43.69/kg) for
certified seed in Danli or a smaller town. At a price of L 12/lb (L 26.4/kg), the estimated
seed demand in El Paraiso and Olancho is of 24 mt and 54 mt, respectively. The total
amount of seed DUWEST is going to buy (23,000 kg) is approximately 30% of the
estimated combined total demand in both departments. This means that DUWEST faces

low risk of not being able to sell its stock of seed.

7.2 Conclusions

In summary, this study suggests that it is feasible to increase small-scale
producers’ access to commercial seed of modern bean varieties by contracting small-scale

farmers to produce the seed and marketing it through an existing input dealer.

First, small-scale farmers and input distributing firms are willing to work together
to increase the supply of improved bean seed and make it more affordable to farmers by
selling commercial seed for L 12/lb (L 26.4/kg)—compared to the price in 2002 of L1 5/lb
(L 33/kg) for commercial seed (Hondugenet), L 15.90/lb (L 35/kg) for certified seed
(Zamorano), and a range from L 17.27 (L 38/kg) to L l9.86/1b (L 43.69/kg) in Danli or a

smaller town.

Second, collaboration with Zamorano for third party certification will establish
trust regarding the quality of seed produced by the small-scale farmers and assures the

willingness of DUWEST to participate in the scheme.

115

Third, a substantial share of small-scale farmers in El Paraiso (35%) and Olancho
(54%) would be willing to pay L12/lb (L 26.4/kg) for commercial seed of the variety
Amadeus 77 due to its preferred seed color« compared to widely grown modern varieties

such as Tio Canela and Dorado.

7.3 Policy Implications

In the past, initiatives to improve small-scale farmers access to improved bean
seed have been largely associated with relief projects. Given the immediate need to make
seed available for the coming planting season, NGO and government projects have
provided farmers free or highly subsidized seed. Clearly, these projects met the
immediate seed needs of farmers. However, they have also undermined the development
of a sustainable seed system. Since private sector seed firms cannot compete against
projects that distributed subsidized seed, the private sector has largely abandoned this

potential market.

Fieldwork for this study was carried out in summer 2002. At that time, it
appeared that it would be possible to design a sustainable bean system and encourage
NGOs and the government to no longer distort demand by distributing subsidized seed.
However, in 2004 DICTA obtained funding from FAQ to initiate a new a seed project.
Under the project, DICTA supplied improved seed to small-scale farmers, who paid the

seed by giving DICTA the same amount of grain of a landrace or traditional varieties or

116

harvest of the modern variety (kilo for kilo”). While the project had a small budget
($16,729.97), it benefitted 600 small-scale bean and corn farmers of the northern
municipalities of Olancho. DICTA distributed the seed to farmers through NGOs, rural
cashiers, and municipalities, which were in charge of collecting grain at harvest as
payment for the seed. In 2005, DICTA submitted a new proposal which was designed to

distribute improved bean and corn seed over a larger region of Honduras.

If the expanded project is funded and implemented throughout the country, the
proposed bean seed scheme can not be implemented, due to the lower cost of the

subsidized seed being offered by DICTA.

Projects like kilo-for-kilo should not be implemented except for relief purposes
because they prevent the development of a private seed market. Private firms cannot

compete with the bean seed prices that such projects offer farmers.

Similarly, NGO-sponsored subsidized seed projects (where food security is the
main goal), that give free seed to farmers prevent the development of a private seed
market. Rather than giving farmers seed, seed schemes should seek to develop linkages
between small-scale bean seed producers and input distributors, as proposed by this study.
For example, NGOs could give poor farmers seed vouchers which they could use to buy
seed at a participating input dealer. Subsequently, the input distributor could redeem the

voucher from the NGO. Such a system will provide farmers more flexibility in their

 

28 This project was based on similar schemes that have been implemented in Mexico,
Guatemala, and Nicaragua. See Gonzalez for an analysis of Mexico’s kilo-for-kilo seed
distribution.

117

choice of seed, enhance competitiveness, and promote the development of a private

sector-based bean seed system (Remington, eta]. , 2002).

Finally, if government is to be involved in bean seed supply, it should direct its
efforts on advertising campaigns to promote the purchase of improved seed and on fairs

that expose farmers to these modern varieties.
7.4 Limitations

The demand results reported in this study apply to only the departments of El
Paraiso and Olancho. While these departments account for a large acreage of bean
planted in Honduras, farmers in other departments may have different varietal preference

and may not be willing to pay a premium for improved bean seed.

Another limitation is that the expected demand for Amadeus will be lower than
projected in this study if Amadeus 77 yields less in farmers fields than Tio Canela and
Dorado. This is because the projected demand for Amadeus 77 was based on the

assumption that Amadeus 77 produced the same yield as those two varieties.
7.5 Future Research

Future research should be directed towards building and supporting stronger
linkages between the input distributing firms and small-scale bean seed farmers (i.e. input

contracting) to insure the sustainability of these types of schemes.

Also, if the proposed scheme is implemented, a follow up study should be carried
out to assess the actual seed demand of small-scale farmers to validate or reject the

demand estimate reported in this thesis. Such a study would provide valuable

118

 

information to seed producers and input distributing firms regarding the actual demand

for improved seed by small-scale farmers.

Finally, a benefit-cost analysis of the current kilo-for-kilo project in Honduras

should be implemented to determine if the project is economically feasible.

119

Appendix A. Explanatory variables correlation matrixa.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

size weight fertilizer age school land Mod Var
size 1.00
weight -0.98 1.00
fertilizer 0.02 -0.02 1 .00
age -0.07 0.08 -0.29 1.00
school -0.01 0.02 0.24 -0.54 1.00
land 0.06 -0.04 0.12 -0.05 -0.05 1.00
Mod Var 0.04 -0.05 -0.09 -0.03 -0.16 0.14 1.00
drcolor 0.00 -0.00 -0.00 0.00 0.00 0.00 -0.00
dlrcolor 0.00 -0.00 -0.00 -0.00 0.00 0.00 0.00
rcolor 0.00 -0.00 0.00 -0.00 0.00 0.00 -0.00
lrcolor 0.00 —0.00 -0.00 0.00 0.00 0.00 -0.00
drcolor dlrcolor rcolor lrcolor
drcolor 1.00
dlrcolor -0.20 1 .00
rcolor -0.20 -0.20 1 .00
lrcolor -0.32 -0.31 -0.31 1.00
a Refer to table 22 for details of variables.
Appendix B. Table of probit predicted outcomes
True
Classified Positive Negative Total
Positive 54 18 72
Negative 101 259 360
Total 155 277 432

 

 

 

 

120

 

 

Appendix C. Probit and truncated model coefficients of bean attributes and farmers’

characteristics on purchase decision and WTP.

 

 

 

 

 

 

 

 

 

 

 

 

Variable Probit Model Truncated Regression
Dark red color -0.007 0.541
Std. error 0.224 1.437
p-value 0.975 0.706
Lighter dark red color 0.277 -0.282
Std. error 0.220 1.287
p-value 0.208 0.826
Red color 1.314 0.637
Std. error 0.228 1.202
p-value 0.000 0.596
Light red color -0.020 0.052
Std. error 0.196 1.299
pvalue 0.91 8 0.968
Fertilizer used in Postrera 2001 -0.150 -1.071
Std. error 0.15] 0.860
-value 0.321 0.213
Age 0.005 0.084
Std. error 0.006 0.036
p-value 0.399 0.02]
Education 0.014 0.354
Std. error 0.019 0.104
p-value 0.451 0.001
Land -0.009 0.149
Std. error 0.007 0.040
p-value 0.201 0.000
Use of modern variety 0.140 2.352
Std. error 0.146 0.916
p-value 0.340 0.010
Constant -0.918 4.454
Std. error 0.447 2.942
-value 0.040 0. 130

 

 

 

121

 

References

Back, E. and Hong, G. (2004) Effects of Family Life-Cycle stages on Consumer
Debts. Journal of Famin and Economic Issues. 25(3): 359-385.

Beatie, BR. and C. R. Taylor (1993) The Economics of Production. Malabar, Florida:
Krieger Publishing.

Bemsten, R. and Mainville D. (1999) Proyectos Artesanales de Produccion de Semilla
en Paises en Desarrollo: Lecciones Aprendidas e Irnplicaciones para el

Disei’io de Esquemas mas Eficaces, in Exmriencias en la Produccion
Artesanalde Semillas de Frijol en Centro America. J .C. Rosas and A. Castro, eds.
Proceedings of Taller de Produccion y Distribucion Artesanal de Semilla dc
Frijol en Centro America. Escuela Agricola Panamericana/ Zamorano, Honduras
(1998)

Burton, M., M. Tomlinson, and T. Young (1994) Consumers’ Decisions Whether or not
to Purchase Meat: A Double Hurdle Analysis of Single Adult households. Journal
of Agp'cultural Economics. 45(2): 202-212.

Chacon, J .R., Hernandez, J.C., and Maroto, MA. (1999) “Produccion Artesanal de
Semilla de Frijol (PASF): Caso de una Asociacion de Productores de la Region
Brunca de Costa Rica, in Exmriencias en la Produccion Artesanal de Semillas
de Frijol en Centro America. J .C. Rosas and A. Castro, eds. Proceedings of
Taller de Produccion y Distribucion Artesanal de Semilla de Frijol en Centro
America. Escuela Agricola Panamericana/ Zamorano, Honduras (1998).

Cragg J .G (1971) Some statistical models for limited dependent variables with
application to the demand of durable goods. Econometrica. 39(5): 829-844.

Douglas, J. (1980) Succesful seed progms: A planning and Managgment Guide.
Boulder, Colorado: Westview Press.

Doye, D. (2005) Budgets: Their Use in Farm Management. Oklahoma Cooperative

Extension Fact Sheet F -l 39. Available at http://www.osuextra.com [accessed
Aug, 28, 2005].

122

Fafchamps, Marcel, dc Janvry, Alain and Elisabeth Sadoulet. (1995) Transaction Costs,
Market Failures, Competitiveness and the State, in G.H. Peters and DD. Hedley
(eds.), Agp'gultural Competitiveness: Market Forces and Policy Choices.
(Proceedings of the Twenty-Second International Conference of Agricultural
Economists, Held at Harare, Zimbabwe, 22-29 August 1994). Aldershot, U.K.:
Dartmouth : 343-54.

Garcia, F. (1999) Experiencias en la Produccion Artesanal de Semilla de F rijol en la

Region IV de Nicaragua, in Exmriencias en la Produccion Artesanal d9 Semillas
de Frijol en Centro America. J.C. Rosas and A. Castro, eds. Proceedings of
Taller de Produccion y Distribucion Artesanal de Semilla de Frijol en Centro

America. Escuela Agricola Panamericana/ Zamorano, Honduras (1998).

Gardner, F.V. (1948) The New Meaning of Breakeven Points, in Production Costs and
Breakeven Points. AMA Production Series, USA.

Hartley, N., Wright, RE, and Adamowicz, V. (1998) Using Choice experiments to Value
the Environment. Environmental Resource Economics. 11(3-4): 413-428.
Hayarni, Y. and K. Otsuka (1993). The Economics of Contract Choice.
Oxford,UK.

Heisey, P.W. and J .P. Brennan (1991) An Analytical Model of F arrners’ Demand for
Replacement Seed, American Journal of Agricultural Economics. 73(4): 1044-52.

Holtzman, J. (1993) Rapid Appraisal Methods in a Diagnostic Assessment of Vegetable
Seed Marketing, in Rapid Appraisal Methods. Eds. Krishna Kumar, The World
Bank. Washington DC.

Jimenez, LA. and M. von Oppcn (1999) The Use of Hedonic Price Analysis in
Agricultural Research: Market Prices and the Quality Characteristics of Beans in

Colombia, in Agp'cultural Markets from Theog to Practice. Barbara Harriss-
White (ed.), New York: St. Martin’s Press.

Kennedy, P. (1998) A Guide to Econometrics. 4‘“ Ed. MIT Press

Karnau, M. (1997). Snap bean seed production and dissemination channels in Kenya.

Network on Bean Research in Africa, Occasional Publications Series, No. 22.
CIAT, Kampala, Uganda.

123

Kumar, K. (1993) An Overview of Rapid Appraisal Methods in Development Settings,
in mid Appraisal Methods. Eds. Krishna Kumar, The World Bank. Washington
DC.

Mather, D. (2003) Essays on the Economic Impact of Disease-Resistant Bean Research in
Honduras. Ph.D. Dissertation. Michigan State University.

Mather, D.; Bemsten,R.; Rosas,J.C.; Ruano, A.V.; Escoto, D. and Martinez, J. (2002)
The Impact of Bean Research in Honduras. Paper presented at the conference

on Impacts of Agricultural Research and Development, San Jose, Costa Rica,
February 4-8, 2002.124

North, D. (1990) Institutions, Institutional Change and Economic Performancg.
Melbourne, Australia. Cambridge University Press.

Nicholson, W. (1998) Miproeconomic Theogy. Forth Worth: The Dryden Press.

Ordofiez, J .L. (1999) Produccion Artesanal dc Semilla de Frijol: Caso Santa
Gertrudis, Quezada, Jutiapa, Guatemala, in Experiencias en la Pm6uccion
Artesanal dc Semillas de Frijol en Centro America. J .C. Rosas and A. Castro,
eds. Proceedings of Taller dc Produccion y Distribucion Artesanal dc Semilla de
Frijol en Centro America. Escuela Agricola Panamericana/ Zamorano, Honduras
(1998)

Perman, R., Ma, Y., McGilvray, J ., and Common, Michael (1999) Natural Resource &
Environmental Economics. 2 Ed. Remington, T., Maroko, J ., Walsh, S., Omanga,
P., and Charles, E. (2002). Getting Off the Seeds-and-Tools Treadmill with CRS
Seed Vouchers and Fairs. Overseas Development Institute. Blackwell
Publishing, Malden, MA.

Remington, T., Maroko, J ., Walsh, 8., Omanga, P., and Charles, E. (2002). Getting Off

the Seeds-and-Tgols Treadmill with CRS Seed Vouchers and Fairs. Overseas
Development Institute. Blackwell Publishing, Malden, MA.

Rodriguez, F. and Viana, A. (1998) Resultados dp Seminario-Taller Acciones Para

Mejorar la Produccion Artesanal dc Semilla de Frijol en Honduras. PROFRIJOL,
Panama.

Roosen, J ., J .A. Fox, D.A. Hennessy, and A. Schreiber (1998) Consumers’ Valuation of

Insecticide Use Restrictions: An Application to Apples, Journal of Agp‘ 'cultural
and Resource Economics. 23(2): 367-384.

124

 

Rusike, J. and Eicher, C. (1997) Institutional Innovations in the Maize Seed Industry, in
Afriga’s Empgging Maize Revglution. D. Byerlee and C. Eicher. Eds. Lynne
Rienner Publishers.

Shaffer, J. (1973). Changing Orientations of Marketing Research. American Journal of
Agp'cultural Economics. 55: 333-35.

Shaffer, J. (1968). On the Concept of Subsector Studies. American Journal of
Agp'cultural Economics 50(5):]437-1449.

Staatz, J .M., (1997) Notes On The Use Of Subsector Analysis As A Diagnostic Tool For
Linking Industry And Agriculture, Staff Papers 97-04, Michigan State University,
Agricultural Economics.

Tripp, R. (1997) The Institutional Conditions for Seed Enterprise Development. Overseas
Development Institute, Working Paper 105, London.

Viana, A. (1999) Esquemas de Produccion Artesanal de Serrrilla de Frijol en Centro
America: Proyecto PROF RIJ OL, in Exxriencias en la Produccion Artesanal dc
Semillas dc Frijol en Centro America. J .C. Rosas and A. Castro, eds.
Proceedings of Taller de Produccion y Distribucion Artesanal dc Semilla de Frijol
en Centro America. Escuela Agricola Panamericana/ Zamorano, Honduras
(1998)

Williamson, Oliver E., (1985) The Economic Institutions of Capitalism, New York: The
Free Press.

125

lItliljjlijljljilljiill