• TABLE OF CONTENTS
HIDE
 Front Cover
 Title Page
 Copyright
 Table of Contents
 Abstract
 Acknowledgement
 List of Tables
 Section
 Acronyms
 Introduction
 Study area
 Maize-based cropping systems in...
 Alternative strategies for intensifying...
 The economics of intensificati...
 Challenges for implementation
 Reference
 Appendix A : Crop budgets
 Appendix B : Labor use
 Appendix C : cost benefit analysis...
 New papers from the Natural Resources...
 Back Cover






Group Title: Paper - Natural Resources Group - 96-07
Title: Intensifying maize-based cropping systems in the Sierra de Santa Marta, Veracruz
CITATION PDF VIEWER PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00077513/00001
 Material Information
Title: Intensifying maize-based cropping systems in the Sierra de Santa Marta, Veracruz
Series Title: Paper (International Maize and Wheat Improvement Center. Natural Resources Group)
Physical Description: vi, 55 p. : ill. ; 28 cm.
Language: English
Creator: Buckles, Daniel, 1955-
Erenstein, Olaf
International Maize and Wheat Improvement Center
Publisher: International Maize and Wheat Improvement Center (CIMMYT)
Place of Publication: Mx́ico D.F. México
Publication Date: 1996
 Subjects
Subject: Cropping systems -- Mexico -- Veracruz   ( lcsh )
Corn -- Yields -- Mexico -- Veracruz   ( lcsh )
Sustainable agriculture -- Mexico -- Veracruz   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Bibliography: Includes bibliographical references (p. 37-38).
Statement of Responsibility: Daniel Buckles and Olaf Erenstein.
General Note: At head of title: CIMMYT.
 Record Information
Bibliographic ID: UF00077513
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 36722777
issn - 1405-2830 ;

Downloads

This item has the following downloads:

nrg96_07 ( PDF )


Table of Contents
    Front Cover
        Front cover
    Title Page
        Page i
    Copyright
        Page ii
    Table of Contents
        Page iii
    Abstract
        Page iv
    Acknowledgement
        Page iv
    List of Tables
        Page v
    Section
        Page v
    Acronyms
        Page vi
    Introduction
        Page 1
    Study area
        Page 2
        Page 3
        Page 4
    Maize-based cropping systems in the Sierra d Santa Marta : The current baseline
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
    Alternative strategies for intensifying maize-based cropping systems
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
    The economics of intensification
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
    Challenges for implementation
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
    Reference
        Page 37
        Page 38
    Appendix A : Crop budgets
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
    Appendix B : Labor use
        Page 48
        Page 49
        Page 50
    Appendix C : cost benefit analysis results
        Page 51
        Page 52
        Page 53
        Page 54
    New papers from the Natural Resources Group
        Page 55
    Back Cover
        Back cover
Full Text



II
CIMMYT
Sustainable Maize
and Wheat Systems
for the Poor


Intensifying Maize-Based

Cropping Systems in the Sierra

de Santa Marta, Veracruz
Daniel Buckles and Olaf Erenstein


Natural Resources Group[


Paper 96-07









II
CIMMYT


Intensifying Maize-Based
Cropping Systems
in the Sierra de Santa Marta, Veracruz
Daniel Buckles and Olaf Erenstein *




Natural Resources Groupo
Paper 96-07






* Daniel Buckles was an anthropologist working with the CIMMYT Economics Program at the
time this research was conducted. Olaf Erenstein is an associate expert with the CIMMYT
Natural Resources Group. The views expressed here are not necessarily those of CIMMYT.



































CIMMYT is an internationally funded, nonprofit scientific research and training organization.
Headquartered in Mexico, the Center works with agricultural research institutions worldwide to develop
sustainable maize and wheat systems for poor farmers in developing countries. It is one of 16 similar centers
supported by the Consultative Group on International Agricultural Research (CGIAR). The CGIAR
comprises some 40 donor countries, international and regional organizations, and private foundations. It is
sponsored by the Food and Agriculture Organization (FAO) of the United Nations, the International Bank
for Reconstruction and Development (World Bank), and the United Nations Development Programme
(UNDP).

Direct support for CIMMYT's research agenda comes through the CGIAR from many sources, including the
governments of Australia, Austria, Belgium, Canada, China, Denmark, France, Germany, India, Iran, Italy,
Japan, the Republic of Korea, Mexico, the Netherlands, Norway, the Philippines, Spain, Switzerland, the
United Kingdom, and the USA, and from the European Union, Ford Foundation, Inter-American
Development Bank, OPEC Fund for International Development, UNDP, and World Bank. CIMMYT also
receives support for complementary research from the International Institute of Tropical Agriculture, the
International Irrigation Management Institute, the Kellogg Foundation, the Rockefeller Foundation, the
Sasakawa Africa Association, and many of the other organizations listed above.

Responsibility for this publication rests solely with CIMMYT.

Printed in Mexico.

Correct citation: Buckles, D., and O. Erenstein. 1996. Int. r, iiiirn: Maize-Based Cropping Systems in the Sierra de
Santa Marta, Veracruz. NRG Paper 96-07. Mexico, D.F.: CIMMYT.

Additional information on CIMMYT is available on the World Wide Webb at:
http: / /www.cimmyt.mx or http: / /www.cgiar.org

ISSN: 1405-2830
AGROVOC descriptors: Mexico; Veracruz; Zea mays; Maize; Sustainability; Plant production; Cropping
systems; Farming systems; Soil conservation; Production economics;
AGRIS category codes: E16; F08
Dewey decimal classification: 338.16











Contents


Page

iv Abstract
iv Acknowledgments
v Tables
v Figures
vi Acronyms
1 Introduction
2 The Study Area
6 Maize-Based Cropping Systems in the Sierra de Santa
Marta: The Projected Baseline
6 The decline of the traditional milpa system
8 The current maize cropping system
10 Major constraints on maize productivity

13 Alternative Strategies for Intensifying Maize-Based
Cropping Systems
13 Cover crops
15 Commercial fertilizers
16 Conservation of crop residues
17 Contoured hedgerows and maize rows
19 Increased maize planting density
19 Other considerations

19 The Economics of Intensification
21 Upland zone: Field-level implications
22 Upland zone: Farm-level implications
26 Lowland zone: Field-level implications
27 Lowland zone: Farm-level implications

31 Challenges for Implementation
35 Conclusions
37 References
39 Appendix A. Crop Budgets
48 Appendix B. Labor Use
51 Appendix C. Cost Benefit Analysis Results










Abstract


This paper focuses on intensifying maize-based cropping systems in the Sierra de Santa Marta
Region of Veracruz, Mexico. Following a description of the study area, the paper examines the
historical forces that have shaped regional maize-based cropping systems, current maize
production practices, and major productivity and sustainability constraints. Potential solutions to
these constraints, some of which have been successfully tested in the area, are described and
possible productivity and sustainability impacts are appraised in qualitative terms. The appraisal
provides the basis for a quantitative analysis of the farm-level costs and benefits associated with
the adoption of the alternative practices. A farmer-based approach to extension is described and
implementation costs are estimated.


Acknowledgments

This paper is a revised version of an earlier study co-funded by CIMMYT and the Global
Environment Facility's (GEF) Program for Measuring Incremental Cost for the Environment
(PRINCE). The original study was undertaken as part of a wider collaborative project by PRINCE,
the Proyecto Sierra de Santa Marta A.C. (PSSM), and CIMMYT; the project attempts to estimate
the incremental cost of biodiversity conservation in the Sierra de Santa Marta area.

The authors would like to acknowledge the valuable comments to earlier drafts of this paper by
Marjatta Eilitta, Gustavo Sain, Karen Dvorak, Raffaello Cervigni, Mario Ramos, Jeff White,
Elizabeth Rice, and Jerome F6urnier.










Abstract


This paper focuses on intensifying maize-based cropping systems in the Sierra de Santa Marta
Region of Veracruz, Mexico. Following a description of the study area, the paper examines the
historical forces that have shaped regional maize-based cropping systems, current maize
production practices, and major productivity and sustainability constraints. Potential solutions to
these constraints, some of which have been successfully tested in the area, are described and
possible productivity and sustainability impacts are appraised in qualitative terms. The appraisal
provides the basis for a quantitative analysis of the farm-level costs and benefits associated with
the adoption of the alternative practices. A farmer-based approach to extension is described and
implementation costs are estimated.


Acknowledgments

This paper is a revised version of an earlier study co-funded by CIMMYT and the Global
Environment Facility's (GEF) Program for Measuring Incremental Cost for the Environment
(PRINCE). The original study was undertaken as part of a wider collaborative project by PRINCE,
the Proyecto Sierra de Santa Marta A.C. (PSSM), and CIMMYT; the project attempts to estimate
the incremental cost of biodiversity conservation in the Sierra de Santa Marta area.

The authors would like to acknowledge the valuable comments to earlier drafts of this paper by
Marjatta Eilitta, Gustavo Sain, Karen Dvorak, Raffaello Cervigni, Mario Ramos, Jeff White,
Elizabeth Rice, and Jerome F6urnier.











Tables


Page


7 Table 1. Deforestation in the Sierra de Santa Marta
34 Table 2. Estimated costs of campaign coordination, 1994
39 Table A-1. Yield adjustment factors
40 Table A-2. Crop budget for the without case in the buffer zone (ha basis)
42 Table A-3. Crop budget for the with case in the buffer zone (ha basis)
44 Table A-4. Crop budget for the without case in the lowland zone (ha basis)
46 Table A-5. Crop budget for the with case in the lowland zone (ha basis)
48 Table B-1. Assumed temporal distribution of activities
49 Table B-2. Labor distribution by year and month on ha basis, buffer zone
49 Table B-3. Labor distribution by year and month on farm basis, buffer zone
50 Table B-4. Labor distribution by year and month on ha basis, lowland zone
50 Table B-5. Labor distribution by year and month on farm basis, lowland zone
51 Table C-1. Farm-level implications over time of intensifying model farm in the buffer
zone (farm basis)
52 Table C-2. Sensitivity analysis of farm-level implications of intensifying model farm in
the buffer zone (farm basis)
53 Table C-3. Farm-level implications over time of intensifying the model farm in the
lowland zone (farm basis)
54 Table C-4. Sensitivity analysis of farm-level implications of intensifying the model farm
in the lowland zone (farm basis)


Figures


Page


2 Figure 1.
8 Figure 2.
11 Figure 3.
23 Figure 4.

28 Figure 5.


Sierra de Santa Marta study area in southern Veracruz, Mexico
Production calendar for Pajpan area, Sierra de Santa Marta, Veracruz
Hypotheses on maize problems and causes, Sierra de Santa Marta, Veracruz
Observed and proposed cropping patterns for the model farm in the buffer
zone, Sierra de Santa Marta, Veracruz
Observed and proposed cropping patterns for the model farm in the lowland
zone, Sierra de Santa Marta, Veracruz











Tables


Page


7 Table 1. Deforestation in the Sierra de Santa Marta
34 Table 2. Estimated costs of campaign coordination, 1994
39 Table A-1. Yield adjustment factors
40 Table A-2. Crop budget for the without case in the buffer zone (ha basis)
42 Table A-3. Crop budget for the with case in the buffer zone (ha basis)
44 Table A-4. Crop budget for the without case in the lowland zone (ha basis)
46 Table A-5. Crop budget for the with case in the lowland zone (ha basis)
48 Table B-1. Assumed temporal distribution of activities
49 Table B-2. Labor distribution by year and month on ha basis, buffer zone
49 Table B-3. Labor distribution by year and month on farm basis, buffer zone
50 Table B-4. Labor distribution by year and month on ha basis, lowland zone
50 Table B-5. Labor distribution by year and month on farm basis, lowland zone
51 Table C-1. Farm-level implications over time of intensifying model farm in the buffer
zone (farm basis)
52 Table C-2. Sensitivity analysis of farm-level implications of intensifying model farm in
the buffer zone (farm basis)
53 Table C-3. Farm-level implications over time of intensifying the model farm in the
lowland zone (farm basis)
54 Table C-4. Sensitivity analysis of farm-level implications of intensifying the model farm
in the lowland zone (farm basis)


Figures


Page


2 Figure 1.
8 Figure 2.
11 Figure 3.
23 Figure 4.

28 Figure 5.


Sierra de Santa Marta study area in southern Veracruz, Mexico
Production calendar for Pajpan area, Sierra de Santa Marta, Veracruz
Hypotheses on maize problems and causes, Sierra de Santa Marta, Veracruz
Observed and proposed cropping patterns for the model farm in the buffer
zone, Sierra de Santa Marta, Veracruz
Observed and proposed cropping patterns for the model farm in the lowland
zone, Sierra de Santa Marta, Veracruz











Acronyms


B / C ratio Benefit-cost ratio
CBA Cost-benefit analysis
CIMMYT Centro Internacional de Mejoramiento de Maiz y Trigo
GEF Global Environment Facility
INI Instituto Nacional Indigenista
INIFAP Instituto Nacional de Investigaciones Forestales y Agropecuarias
LUI Land use intensity
NGO Non-Governmental Organization
PRINCE Program for Measuring Incremental Cost for the Environment
PSSM Proyecto Sierra de Santa Marta, A.C.
SAGAR Secretarfa de Agricultura, Ganaderfa y Desarrollo Rural
SARH Secretarfa de Agricultura y Recursos Hidrdulicos (now SAGAR)
SEDAP Secretarfa de Desarrollo Agropecuario y Pesquero, Gobierno del Estado, Veracruz
SEDESOL Secretarfa de Desarrollo Social











Intensifying Maize-Based Cropping Systems
in the Sierra de Santa Marta, Veracruz


Daniel Buckles and Olaf Erenstein


Introduction

Farmers in the Sierra de Santa Marta, an
indigenous region of southern Veracruz,
Mexico, have met their subsistence needs for
generations by cultivating the lower slopes and
adjacent hillsides of the Sierra using relatively
land-extensive cropping practices. While these
practices were viable and relatively sustainable
in the past, they are no longer so due to the
depletion of forested lands and increasing
land-use pressures in areas already opened for
agriculture. Intensification has lead to soil
erosion and declines in soil fertility, with a
parallel decline in agricultural productivity and
forest cover. Technical options exist, however,
that can potentially intensify the current
system in a sustainable and equitable manner.
Cover crops, soil conservation practices, and
moderate amounts of external inputs can
increase the efficient use of existing resources
while maintaining or improving the resource
base. This paper provides a detailed review of
this option for intensifying maize-based
cropping systems in the Sierra de Santa Marta,
with special emphasis on the farm-level costs
and benefits. Means of facilitating farmer
access to and adaptation of these practices are
also considered.

The paper is part of a broader collaborative
study initiated by the Program for Measuring
Incremental Cost for the Environment
(PRINCE), a research program of the Global
Environmental Facility (GEF) that is


developing a set of practical tools for
estimating the incremental cost1 of actions that
protect the global environment (PRINCE 1994).
The study is an attempt to estimate the
incremental cost of biodiversity conservation in
the Sierra de Santa Marta area and is being
undertaken by PRINCE, the Proyecto Sierra de
Santa Marta A.C. (PSSM), and CIMMYT. This
paper focuses on maize-based cropping
systems in the Sierra de Santa Marta; it is
intended to complement studies of other
sectors in the area (GEF/PSSM/CIMMYT,
forthcoming).

Following a description of the study area, the
paper examines the historical forces that have
shaped maize-based cropping systems in the
Sierra de Santa Marta, current maize
production practices, and major productivity
and sustainability constraints. Potential
solutions to these constraints, some of which
have been successfully tested in the area, are
described and possible productivity and
sustainability impacts are appraised in
qualitative terms. The appraisal provides the
basis for an analysis of the farm-level costs and
benefits associated with adoption of the
alternative practices. A farmer-based approach
to extension is described and implementation
costs are estimated, followed by a summary of
the study's main conclusions.



1 "Incremental cost" refers to the additional economic
burden that developing countries have to bear when
they take actions that benefit the global environment
and that go beyond national development goals
(PRINCE 1994).











The Study Area


The Sierra de Santa Marta is a remote mountain
range in southern Veracruz located on the gulf
coast between 180 15'N and 180 30'N latitude
(Figure 1). It encompasses some 1,200 km2
rising steeply from sea level at the Gulf of
Mexico to more than 1,700 meters at its highest
peak. Three municipalities-Pajapan, Soteapan,
and Mecayapan-make up the northeastern
and southern slopes of the range, while
portions of two other municipalities-
Catemaco and Hueyapan de Ocampo-
comprise much of the western slope.

The Sierra is part of the Olmec heartland, a
region where Mesoamerica's mother culture
developed between 1200 and 400 B.C. (Garcia
de Le6n 1976, p. 280). The population is mainly
indigenous Nahua and Popoluca speakers that
have inhabited the area since pre-Hispanic
times. Mestizo communities of more recent
origin are located at various points throughout


the Sierra, although mainly on the northern
slope where ranching activities dominate. The
Sierra population amounts to approximately
40,000 people in 91 communities, most of which
are ejidos. Agriculture, cattle ranching, coffee
production, fishing, and the extraction of forest
resources are the main economic activities of the
population, but incomes are inadequate. Some
80% of the population lives in extreme poverty
with incomes of less than US$ 1,200 annually
(Arias 1991). The Sierra is one of the poorest
regions in Veracruz and among the poorest in
the nation.

To the south and east of the Sierra de Santa
Marta lie the cities of Coatzacoalcos, Minatitlan,
and Jaltipan-which have a total population of
over one million. These cities comprise Mexico's
most important petro-chemical complex. To the
southwest lies the bustling cattle-ranching and
commercial city of Acayucan. A major center for
the regional cattle industry, the city exerts an
important influence on developments in the
Sierra. To the west lies the
Tuxtlas region, which is
important to the state economy
as a center of tobacco
production, ranching, and
commercial maize.
Golfo de
'tde Mxco The sudden rise of the Sierra
lexico eracru
de Santa Marta from sea level
to over 1,700 meters creates a
wide range of climatic
/% conditions. The moisture-laden
0+o trade winds from the Gulf of
Mexico are transformed into
a Vi er over 4,000 mm of rain on the
Tabaso northern slope, but as little as
Coatzacoalcos
lc 1,200 mm fall on the

Rio
"coatza-a'cos Figure 1. Sierra de Santa Marta
study area in southern
Veracruz, Mexico.










southwestern slope due to a rain shadow effect.
Still, rainfall is generally abundant throughout
the region and distributed in a bimodal pattern
which allows for a growing season of more than
270 days. The altitudinal gradient modifies the
temperature regime, and the soils, although
generally of volcanic origin, are now quite
diverse due to both climatic factors and varied
land uses. The flora and fauna are also
extremely diverse: some 3,000 plant species and
1,149 animal species have been documented, a
number of which are unique to the region
(Ramirez 1991). More than 450 bird species,
almost 40% of the known bird species in Mexico,
frequent the forests of the Sierra de Santa Marta
and neighboring Tuxtlas range (Schaldach,
Escalante, and Winker forthcoming). These
unique features have prompted various formal
declarations aimed at conservation, including
the creation in 1980 of a Special Biosphere
Reserve. Little action has been taken, however,
to conserve the remaining 20,000 or so ha of
forested land in the region (Par6 et. al.,
forthcoming; GEF / PSSM / CIMMYT,
forthcoming; Chevalier and Buckles 1995).

Three main agro-ecological zones can be
identified.

* Forest zone: This refers to mountain peaks and
craters (between 1,200 to 1,700 meters) where
most of the remaining 20,000 ha of forest
resources are concentrated. This zone is
generally uninhabited, although frequently
used by people from the upland zone.

* Upland zone: This refers the higher slopes
(between 800 and 1,200 meters), and
comprises communities bordering directly on
the forest. Inhabitants make frequent use of
forest resources and continue to open the
forest margin.


* Lowland zone: This refers to the lower slopes
and sloping land from sea level to about 800
meters, comprising the largest portion of the
population and oldest communities in the
region. Historically, the population of this zone
has had considerable impact on regional
forests; at present that impact has been
reduced.

The analyses presented in this paper are based on
these zones.2

The forest zone is covered with various kinds of
montane forest of considerable biological
importance (Ramirez 1991; Ramirez 1984; Andrle
1964). In this zone, average annual temperatures
are 180 C or less, and rain falls in excess of 4,000
mm per year, providing the cool, wet conditions
needed for development of these forest types. The
mountain peaks and craters are frequently
shrouded in clouds, adding through condensation
as much water to the regional hydrology as falls
in rain (Ramirez 1991). This water is gradually
released to lower areas throughout the year,
supplying the major bodies of water in the region,
including the coastal lagoons Sontecomapan and
Osti6n, Lake Catemaco (Mexico's third largest),
and numerous rivers feeding both the
Coatzacoalcos and Papaloapan watersheds. Some
80% of Coatzacoalcos' potable water, 20% of the
water consumed by MinatitlAn, and virtually all
of the drinking water of Acayucan is drawn from
two springs on the southern slope of the Sierra,
serving more than two million people. The
agricultural potential of this zone is very limited,
mainly because slopes exceed 60%, although the
soils are relatively fertile Andosols.

The upland zone of the Sierra is characterized by
various kinds of subtropical rain forest as well as
oak and pine forests, particularly on the southern
slope. Maize and coffee production are the most


2 For more details on the zones and the agroecological features of sub-zones, see Ramirez et al. 1995 and Par6 et al. 1993.










important agricultural land uses in most of this
zone, although in a few Mestizo communities
located on the northern slope of the upland zone
pastures are dominant. Rainfall in this zone is
high, ranging from 2,000 to 3,000 mm over most
of the zone, but strong winds limit agricultural
potential. Nortes peaking at over 100 km/hr
strike the region from September through
March, lodging maize fields and deflowering
fruit trees and bean crops. Hot, dry winds
known as suradas can also devastate crops
between February and May. These strong winds
confine most forms of agricultural production to
the main rainy season (temporal) from June
through September. The soils of the upland
zone, composed mainly of Andosols, are slightly
acid (pH 6.0) and extremely poor in
phosphorous (0.6 ppm, Bray) (Tasistro 1994).
Levels of organic matter (4.0%) are very high, in
keeping with the relatively extensive land use
patterns still present in this zone (Tasistro 1994).
However, slopes of 30-60% in agricultural areas
make these soils very susceptible to erosion.

The lowland zone was once covered with tropical
rain forest, virtually all of which has given way
to pastures, crops, and secondary vegetation.
The relative importance of ranching and crops
varies considerably within the zone. In general,
however, the northern, southern, and eastern
slopes of the lowland zone are dominated by
pastures; annual cropping is the main land use
on the western slope. Rainfall varies
considerably, although it is generally
distributed bimodally. Two growing seasons
are usually possible, a main season (temporal)
from June through September and a minor


season (tapachole) from October through March.
A short dry season that interrupts all
agricultural activity occurs later as hot air
masses (suradas) sweep in from the south from
March through May. These dry winds pose
considerable risks to crops like maize that are
planted during the minor season (see below).
The soils of the lowland zone are mainly
Luvisols and Vertisols. They exhibit a
moderately acid condition (pH 5.5-6.0) and are
low in P (0.5-1.5 ppm, Bray) as well as other
base nutrients (K, Ca, Mg) (Tasistro 1994). In
general, the cation exchange capacity of the
subsoil in both the upland and lowland zones is
very low, a condition that reduces the capacity
of the soil to retain nutrients and increases the
vulnerability of crops to the effects of soil
erosion and drought (Tasistro 1994).

Most communities in the Sierra are ejidos,
comprising a specific territory and an
association of producers (ejidatarios) with rights
to use the collectively owned land. All but seven
ejidos in the Sierra have been formally
subdivided into individual parcels of an equal
size distributed among ejido members.3 Ejidos
vary considerably in total size and membership
as well as in the amount of land available to
individual ejidatarios. The largest ejido
(Pajapan4) comprises almost 14,000 ha and has
over 900 members. Most ejidos are much smaller,
however, with membership of 100 ejidatarios or
fewer. Individual parcels range from as small as
4 hectares in some ejidos to as large as 25
hectares in others. All ejidos retain some lands,
typically forests, for communal use.


3 Non-parceled ejidos include San Fernando, Ocotal Grande, Ocotal Chico and El Tulin in the municipality of Soteapan;
Santa Rosa Loma Larga in the municipality of Hueyapan de Ocampo; Plan Agrario in the municipality of
Mecayapan; and El Pescador in the municipality of Pajapan. In these ejidos, access to specific parcels of land is
regulated by customary land use rights.
4 Technically, Pajapan, the capital of the municipality of the same name, is an agrarian community (comunidad agraria),
not an ejido. Agrarian communities have a particular form of land tenure reserved for indigenous populations that
chose to exercise traditional claims to land rather than request state land grants in the form of ejidos. While formally
different land tenure systems, agrarian communities and ejidos typically function in much the same way.










Recent revisions to Mexico's constitution
(Article 27) allow ejidatarios to legally rent, lease,
or sell individual ejido land rights, with
permission from the general assembly of
ejidatarios. This reform has increased pressure to
subdivide the few remaining collectively held
ejidos, creating considerable uncertainty and
conflict in a number of communities. The
process of parceling has also constrained the
adoption of land-conserving technologies such
as living fences, which require secure land
tenure (see below).

Other forms of land tenure within the Sierra
include associations of independent producers
coloniess), which are mainly concentrated on the
northern slopes of the Sierra. These land
holdings are considerably larger, 50 ha per
colono in the case of Perla del Golfo. Very small
extensions of private titled property (pL 1i/L fi7
propiedad) are located on the western slope, and
large blocks of state lands are located in the
forest zone. These later lands form the core of
the Special Biosphere Reserve.

Each ejido has a town site, an area specifically
designated for habitation. In general, such town
sites are also used by residents known as
avecindados, usually family members of
ejidatarios with no land-use rights of their own.
These families may gain access to farmland held
by family members in exchange for labor, a
share of the harvest, or cash. In non-parceled
ejidos, residents may also gain access to vacant
land over which they acquire squatters' rights
over time. In many communities, however,
access to land is highly unequal. While data on
the distribution of land for the Sierra as a whole
are limited, evidence from specific communities
highlights problems of land access found
throughout the region. For example, Pajapan is
a lowland zone community of approximate


9,000 people, 40% of whom are landless
(Chevalier and Buckles 1995). In many of the
upland zone communities, the number of
landless avecindados is greater than the number
of ejidatarios.

While the distribution of agricultural land is
skewed, even landless households engage in
some agricultural production, usually maize,
for subsistence purposes. More than 95% of the
households in the upland zone grow some
maize (Rice, Godinez, and Erenstein,
forthcoming). In the lowland zone,
occupational opportunities are more diverse,
making it possible for some families to meet
their maize requirements through cash
purchases. Nevertheless, even in this zone, an
estimated 80% of the households grow some
maize for subsistence purposes (Chevalier and
Buckles 1995). It is important to note, however,
that the degree of self-sufficiency in maize in
both zones is highly variable; in Pajapan, for
example, most households purchase maize for
household consumption during at least four
months of the year. The loss of self-sufficiency
in maize production throughout the region is
due in part to the decline of the traditional
milpa system, as discussed below.


Maize-based Cropping Systems in
the Sierra de Santa Marta:
The Current Baseline

This section examines the historical factors that
have affected the maize-based cropping
systems in the Sierra de Santa Marta and
provides an outline of the current maize
production practices. The major problems
currently affecting maize producers in the
region are summarized; these are problems that
will likely prevail in the near future and may
worsen if no action is taken.










The decline of the traditional
milpa system
Using slash-and-burn techniques, the
indigenous population of the Sierra de Santa
Marta has been growing maize for more than
4,000 years (Stuart 1978; Foster 1942; Garcia de
Le6n 1976). It is the main crop in a diversified
cropping system known as a milpa. In this
system, forest land is cleared, cultivated with a
wide range of crops for a few years, and
abandoned to natural regrowth. A new milpa is
then established on forested land, where the
cycle is repeated.

Land management in the traditional milpa
system of the Sierra de Santa Marta was
circular: farmers "rested" a field when the land
became "tired" and returned to it when
agricultural potential had been restored under
secondary vegetation. Farmers managed
secondary vegetation as a future milpa; they
noted subtle changes in soil characteristics,
weed populations, and plant species-all of
which were indicators of soil fertility.
Secondary vegetation, known as an acaual, was
preferred over mature forest because it
required much less time to clear and was
equally productive (Stuart 1978). Milpas were
surrounded by acauales at various stages of
regrowth and by areas of mature forest from
which natural vegetative succession could
occur. Thus, the traditional milpa system of the
Sierra de Santa Marta was a forest-linked
system based on the shifting of fields and the
constant regeneration of forest species.

Fallow successions were complemented by a
multiple cropping strategy. The traditional
milpa of the Sierra de Santa Marta was a field of
maize widely planted in rows and interseeded
with 10 to 20 other crops at various times of the
year (Foster 1942; Stuart 1978; Perales 1992).
Most milpas contained climbing beans, squash,
sweet potatoes, pigeon peas, sesame, yam bean,


dasheen, cherry tomatoes, chiles, and other
plants intercropped or volunteering among the
maize plants. In addition, sugar cane, plantains,
cassava, pineapple, and papaya were planted
in a section or along the border of the milpa.
While most crops other than maize were grown
in very small quantities, they were important to
family nutrition and presented a number of
ecological advantages (Stuart 1978).

The key to milpa cultivation in the Sierra de
Santa Marta as elsewhere is the length of the
cropping and fallow periods. Continuous
cultivation of tropical soils leads to a decline in
yields and an increase in weeds, the general
reasons for field shifting (Weischet and
Caviedes 1993). The time required to fully
restore the agricultural potential exhausted by
cultivation varies greatly from one area to
another depending upon rainfall, plant species
composition in the secondary vegetation, soil
type, the period of cultivation, and farming
techniques. Scientists generally hold, however,
that fallow periods must exceed cropping
periods if long-term soil impoverishment is to
be avoided (Sanchez and Cochrane 1980;
Weischet and Caviedes 1993).

In most of the lowland zone of the Sierra, two-
year cropping periods (3-4 cycles) were
traditionally followed by approximately eight
years of fallow (Chevalier and Buckles 1995;
Perales 1992). Farmers report that after rested
lands are continuously cropped for 3-4 cycles,
yields decline to less than 800 kg/ha and
weeding time doubles. Fallowing for eight
years was enough to eradicate grassy weeds
and regain the agricultural potential of the
land. Stuart (1978) reports that along the
northern coast of the Sierra where rainfall is
higher, cropping periods of 3-5 cycles were
traditionally followed by a fallow period of
equal duration, with no apparent negative
effects on system sustainability. As Stuart










reports, "No informants reported having used
longer fallow periods in the past, nor do they
consider fallows longer than five years to be in
any way superior to present fallowing periods"
(1978, p. 315). On the drier eastern slope of the
Sierra (e.g., in Soteapan), traditional fallow
periods were somewhat longer, perhaps as
much as 10 years on average (Perales 1992).

Soil erosion caused by traditional milpa
practices was probably quite acute on steep
slopes due to the exposure of the soil after crop
residues are burned. Lands with minor slopes
may not have suffered significantly from soil
erosion, in part because even after burning,
fields were littered with large trunks which
would break the force of rain runoff. Stuart
(1978, p. 311) notes that the land cultivated
along the northern coast of the Sierra using
traditional techniques did not suffer from soil
erosion or exhibit other evidence of land
degradation.

Under conditions of low population density
and limited alternative land uses, the
traditional milpa system satisfied the
subsistence needs of regional farmers and
produced small amounts of surplus production
for sale at regional markets (mainly beans and
pigs fattened with surplus maize). Agriculture
co-existed with the forest, and land
degradation was probably limited. These
conditions no longer hold, however. Beginning
in the 1960s, a land-hungry cattle industry on
the rise displaced farmers from traditional
farming areas and converted secondary
vegetation and forest into pasture. According
to census data from the Secretary of
Agriculture and Water Resources (SARH), the
area under pasture in the three main
municipalities of the Sierra (Pajapan,
Mecayapan, and Soteapan) increased from 19%
in 1950 to 49% in 1988, while agricultural land
uses remained stagnant or declined.


Meanwhile, the area under forest decreased
dramatically: some 75% of the original forest
was lost in 20 years (Table 1).

The conversion of secondary vegetation and
forest into pasture was accompanied by a
process of land concentration. Regional and
local ranchers displaced farmers from the best
lands in the lowland zone through the
manipulation of communal land-tenure
systems, coercion, land purchases, and various
other forms of economic and political power. In
one large community (Pajapan), some 2% of the
population (ranchers) controlled more than half
of the community land (Chevalier and Buckles
1995). Landless farmers were forced into the
forest margin where new communities were
established in more remote areas of the lowland
zone and in the upland zone, thereby extending
the agricultural frontier. More frequently,
however, farmers were pushed out of
agriculture altogether into marginal urban
employment in the industrial centers of
Coatzacoalcos and Minatitlin. The municipality
of Pajapan experienced a net population loss
during periods of rapid growth in the cattle
industry due to out-migration by displaced
farmers (Chevalier and Buckles 1995).

Expansion of the cattle industry displaced
forests, but it also forced farmers to cultivate
those areas open to agriculture more
intensively, with grave implications for
agricultural resources and system performance.


Table 1. Deforestation in the Sierra de Santa Marta

Forest Cumulative forest
Year (ha) loss (%)

1967 81,170 0
1976 55,190 32
1986 21,170 74
1990 20,000 76

Source: Ramirez (1992).











In the lowland zone, fallows of only two years
are now common, while in the upland zone
fallow periods have been reduced to four years
or less. While the relatively longer fallows in
the upland zone probably result in higher
levels of soil fertility, in both zones the current
maize cropping system lacks many of the
adaptive features of the traditional milpa, as
discussed below.

The current maize cropping system
Maize is still the most important crop grown in
the Sierra, and while other food crops may
play a minor role in the cropping system, the
requirements of maize set the tone for all
agricultural activities (Figure 2). In the lowland
zone, climatic conditions facilitate two maize
seasons per year, the temporal or summer
season and the tapachole or winter season. In
the upland zone, summer maize is the only
significant season due to the low productivity


Dry season
---------I


of the winter cycle (see below). Otherwise,
maize cropping practices are similar for
both zones.

Planted in June and harvested between
November and January, the temporal maize
crop is the main one of the year. Land
preparations begin during the winter season
when fallow vegetation (acaual) is cut down by
hand with a machete and burned in the field
once it has thoroughly dried. The soil is
typically not tilled. Mechanized or animal
traction is used by fewer than 5% of the
farming population in the lowland zone and
not at all in the upland zone, where steeply
sloping fields prohibit this form of land
preparation.

Once the field is cleared, a dibble stick is used
to punch a hole in the ground into which maize
seeds are placed. Since moisture is critical to


Wet season


Cool, wet season
t-------------


Type of
production Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Land clearing
Forest -5
Acahual
Maize ae4(
Maize Planing Weediig Farvest k
. Wet season
SWen eding Harves / Flanting -3(
-= Dry season
< Bush beans Pi rves 2(
SPlnting planting Harvest
Manioc Har est
Har est F nting 1(
Chayote Plan 1
--- ^ ^ ^ Plantin,
Fruit trees
S_ Ha est
Burn pasture
S Repair fencing ........ ..... -----------------
Breed animals ----- ----------- --------- ----------------
Sell steers ------ ------------___ ___ ___ --__ __-------

Figure 2. Production calendar for Pajapan area, Sierra de Santa Marta, Veracruz.
(Key: Solid horizontal lines indicate regular activity; broken horizontal lines irregular activity)


00

00


00


00 3-

00










the germination of maize seed, most farmers
plant their maize when they feel certain that
the summer rains have begun in earnest,
typically by early June. While premature
planting can result in crop failure due to poor
germination, planting too late increases the risk
of damage to the crop caused by strong winds
(nortes) late in the season.

Maize in the Sierra de Santa Marta is typically
planted in rows one meter apart. Hills within
rows are also a meter apart, with three to four
seeds in each hill, resulting in a density of
approximately 30,000 to 40,000 plants per
hectare. Perales (1992) reports that farmers
have replaced very tall traditional maize
varieties with shorter varieties derived from
the same landraces (Tuxpeno and Olotillo) to
reduce the incidence of lodging caused by
strong winds. Traditionally, farmers grew
white, yellow, black, and red maize types. Now
white maize has largely replaced other maize
types in the lowland zone. Both white and
yellow maize are still grown in the upland
zone. Very small areas of black and red maize
(Perales 1992; Blanco 1995) are still grown in
both zones. Current local maize varieties
require approximately 100 days to attain
physiological maturity in the lowland zone
and, due to lower temperatures, slightly more
in the upland zone (120 days). The genetic
potential of local maize varieties is generally
high, although additional yield gains can be
made using modern varieties under farmer
conditions (Blanco, Buckles, and Perales 1994).

Being very susceptible to weed competition,
maize requires relatively clear cultivation,
especially during the early stages of plant
development. Maize is typically weeded twice
during the summer season, in early July and in


August. According to a recent survey, an
estimated 92% of the regional farming
population currently use some herbicides to
control weeds (Buckles 1995). The contact
herbicide Paraquat is the most common,
sprayed directly on the weeds between rows of
maize; it is applied with a back-pack sprayer.5
The main reason why farmers have turned
increasingly to herbicides is that manual
weeding with a machete or hoe is the most
time-consuming of farming operations.

Fertilizer use in the Sierra de Santa Marta is
very recent and subject to periodic fluctuations
in cost and availability. Two-thirds of the
farmers who reported using commercial
fertilizer in 1994 began using the input less
than four years earlier, typically as a result of
agricultural credit programs introduced by the
National Indigenous Institute (INI) and the
national social program SEDESOL (Buckles
1995). Perhaps as many as 50% of the farmers
in the region were using small amounts of
fertilizer on their maize in the early 1990s, but
by 1995 this had dropped to less than 20% due
to suspension of the credit programs, increases
in the price of fertilizer, and uncertainties in
maize markets created by the most recent
Mexican economic crisis.

Depending upon the time of planting, the first
tender ears of maize can be harvested
beginning in late August. Maize plants are
doubled below the main ear after reaching
physiological maturity and left in the field to
dry thoroughly. The maize cobs are harvested
piecemeal throughout the winter, even as late
as March. The harvest is carried manually or by
horse from the field to the home. Maize is
stored on the cob, in the kitchen loft (tapanco),
or in stacks on the house floor. Summer maize


5 Paraquat is a highly toxic herbicide, and has caused numerous cases of toxic shock among farmers in the Sierra de
Santa Marta. It is preferred by farmers over other herbicides available in the region due to its lower price.










yields currently average around 1.2 t/ha in the
lowland zone and about 1 t/ha in the upland
zone (Chevalier and Buckles 1995; Perales 1992;
Rice, Godinez, and Erenstein, forthcoming). The
relatively higher average yields in the lowland
zone, despite shorter fallow periods, are due to
the higher incidence of fertilizer use.

In the upland zone, the summer season temporal
is the only significant maize season. In the
lowland zone, however, periodic rainfall from
November to February makes it possible to
grow winter maize (tapachole), provided that
certain modifications are made to the cropping
system. Winter maize is usually planted in
November between the rows of doubled
summer maize, a task eased by the presence of
the earlier maize rows. Weed and crop residues
from the summer season are not burned prior
to planting winter maize but rather are left on
the field as mulch to conserve soil moisture.
Because of the relatively dry winter conditions,
maize plants do not need to be doubled prior to
harvest and only one weeding is required; these
advantages reduce labor costs as compared to
costs in the summer season. Nevertheless, the
risk of crop failure due to drought stress during
the later part of the season is high, and maize
crops can be flattened by strong winds during
the early part of the season. Birds are more
problematic during this season as well. As a
result of these constraints, the most important
of which is drought stress, winter yields are
lower than summer yields, averaging only 500
kg/ha throughout the lowland zone.

Although traditionally maize yields were also
complemented by the harvest of other food
crops in the milpa, most current fields consist
primarily of maize and do not yield significant
intercrops. The use of herbicides has made it
more difficult to manage intercrops and


volunteer food plants, and the practice has thus
declined. Climbing beans traditionally
interseeded in maize have been replaced by
small areas of bush beans grown as sole crops
in a separate part of the field. Other food crops
such as plantains, cassava, and fruit trees are
also absent from many farmers' fields (Perales
1992; Chevalier and Buckles 1995).

The total land area dedicated to maize
production per household and the frequency of
winter maize production have also been
modified by increasing land pressures.
Traditionally, farmers in the Sierra de Santa
Marta made summer milpas of approximately
two hectares and winter milpas of a hectare or
so (Blom and La Farge 1926; Foster 1942; Beaz-
Jorge 1973; Stuart 1978). Currently, the total
maize area is smaller, and not all farmers
cultivate winter maize. In the lowland zone,
farmers cultivate on average only 1.3 hectares of
summer maize and 0.6 hectares of winter maize
(Chevalier and Buckles 1995; Perales 1992;
Buckles and Arteaga 1994). Not all farmers in
the lowland zone can grow winter maize due to
constraints on the availability of suitable land.
In the upland zone, farmers currently cultivate
on average two hectares of maize,
predominantly in the summer season (Rice,
Godinez, and Erenstein, forthcoming). As noted
above, growing winter maize in the upland
zone is a high-risk activity due to strong winds
and erratic rainfall.

Major constraints on maize productivity
Two main categories of problems constrain the
productivity of the current maize-based
cropping system in the Sierra de Santa Marta:
problems associated with the degradation of
fallow land and problems resulting from
inadequate adjustments by farmers to new
circumstances (Figure 3).










Reduced fallow periods. Increasing land use
intensity and corresponding reductions in
fallow periods have resulted in incomplete
regeneration of the fallow vegetation essential
to the recovery of agricultural potential in
shifting cultivation systems. Fallow periods of
two years are now common in the lowland
zone, and these fallows consist entirely of grass
species rather than the woody tree species,
shrubs, vines, and herbaceous plants
characteristic of secondary forest. Slashing and
burning such grassy fallows cannot support
crop production without drawing heavily on
the already limited soil resources. Soils subject
to frequent cultivation are consequently
depleted of fertility, resulting in poor crop
yields and soil chemical imbalances as some
nutrients are mined from the soil. Land
degradation is probably not as severe in the
upland zone due to the longer fallow periods.


Shorter fallow periods and the consequent
elimination of tree species during this period
have also contributed to the build-up of grassy
weeds in many farmers' fields. Stuart (1978)
reports that weeds were not considered a major
problem in the traditional milpa system. By
contrast, weed control in maize currently
represents the single most important cost of
production throughout the lowland zone (see
economic analysis below). Chemical weed
control has increased in recent years, but farmers
have not yet developed sufficient knowledge of
safe management practices. Paraquat, the most
commonly used herbicide in the Sierra, is a
highly toxic chemical that in recent years has
reportedly caused several deaths from toxic
poisoning and numerous cases of toxic shock
due to improper handling. The practice of
washing back-pack sprayers in streams and
rivers may also have affected aquatic life.


Figure 3. Hypotheses on maize problems and causes, Sierra de Santa Marta, Veracruz.
(Key: problems are indicated by rectangles; causes by hexagons; outcomes by diamonds; and adaptations to
outcomes by ovals)










Frequent field burning. As noted above,
slashed secondary vegetation and crop residues
were burned in the traditional system to clear
the field for planting, kill weed seeds, reduce
the incidence of maize diseases (e.g., black spot,
chahuistle) and pests such as rats, and convert
vegetation into ash available for cultivated
crops. While this strategy is well adapted to the
management of mature fallows, it is much less
effective in systems where fallow periods are
too short for significant regrowth. Fallow fields
throughout the Sierra are no longer composed
of large trees, vines, and herbaceous growth
but rather grasses and small shrubs that
present few obstacles to planting. Burning
these residues leaves the soil almost bare,
which in turn increases the risk of soil erosion
during the heavy rains early in the season.
Furthermore, the use of fire to manage short
fallows breaks the plant-to-plant recycling
process characteristic of traditional shifting
cultivation systems; only very little ash is left
on the field to nourish maize plants after
annual weed and crop residues are burned.
Finally, the loss of soil organic matter may also
reduce the moisture-holding capacity of the
soil, with negative impacts on the potential of
land to support winter maize production (not
included in Figure 3). In short, under
prevailing fallow practices, burning provides
few benefits and imposes new costs on the
cropping system.

Row planting down slopes. Planting maize in
rows up and down the slope facilitated the
manual weeding with a hoe of steep hillsides in
the traditional system. However, the practice
also favors soil erosion, especially in the
current situation where runoff is relatively
unimpeded by the intercrops, tree trunks, and
roots that littered traditional milpa sites. While
no formal studies of soil erosion have been


conducted, field observations suggest that both
sheet and channel erosion have been severe; the
causes include inappropriate planting
arrangements, hand weeding in rows, and
burning of crop residues (Tasistro 1994;
Chevalier and Buckles 1995; Perales 1992;
Gutierrez 1995).

Low plant density. As noted above, maize
plants were widely spaced in the traditional
system to accommodate intercrops and very
tall maize varieties. Low plant densities have
been retained, however, despite the dramatic
reduction in intercropping and a tendency for
local varieties to be shorter. This practice limits
the maize yield potential per unit of land and
leaves the soil relatively unprotected from
erosive rainfall. While low soil fertility and
limited use of commercial fertilizers may partly
explain farmer reluctance to increase plant
densities, preliminary adaptive research
suggests that such increases would improve
maize productivity, at least in the more fertile
soils (Blanco, Buckles, and Perales 1994).

The reduction of fallow periods resulting in
low soil fertility and weed invasion is the most
important problem currently constraining
maize productivity in the Sierra de Santa
Marta. Longer-term problems of sustainability
in the milpa system include frequent field
burning and row planting down sloping land-
traditional practices which have become
inappropriate under the current circumstances.
Low plant density also constrains the
productivity of the maize system. Other
problems such as inefficient use of commercial
fertilizers, reduced biological control of pests in
monocropped maize, and inadequate post-
harvest maize storage (not discussed) are also
present in the region, although such problems
are of lesser overall importance.










Alternative Strategies for
Intensifying Maize-Based
Cropping Systems

Effective responses to major constraints on
maize productivity have been developed by
regional farmers and adapted to more intensive
conditions through participatory research by
farmers and scientists, as reported below. The
negative effects of reduced fallows on soil
fertility and weed populations can be
compensated for, in part, through the use of
cover crops and moderate amounts of external
inputs (fertilizer and herbicide). Further farmer
adjustment to new farming conditions can be
accelerated by improved farmer access to
appropriate information regarding
technological options and local adaptive
research, extension, and incentive schemes. Soil
conservation techniques involving the use of
contoured hedgerows, contour planting of
maize, and the conservation of crop residues
are among the most promising means of
reducing erosion in the current system and
enhancing the sustainability of maize
production. Adaptive research and extension
focused on optimal maize planting densities
and the improvement and management of local
maize varieties may also help mitigate current
problems and enhance the productivity and
sustainability of the system. This section
describes these technological options and
assesses their productivity and sustainability
impacts in qualitative terms. The subsequent
section estimates the farm-level costs and
benefits associated with the adoption of these
alternative practices.

Cover crops
The advantages of using cover crops in tropical
agriculture have been widely recognized and
documented (Giller and Wilson 1991; Wade
and Sanchez 1983). These crops, grown in
association with food crops, can be efficient


sources of N, improve soil physical and
chemical properties, control pests and weeds,
and reduce erosion. Cover crops can also
supply food, feed, and fuel. Since the 1950s, a
few farmers in the Sierra de Santa Marta have
been using Mucuna spp. (aggressive, leafy
legumes) in their maize fields to improve soil
fertility and eradicate weeds (Buckles and
Perales 1995). Farmers in San Pedro Soteapan,
the cultural center of the Sierra Popoluca, have
indicated that they encountered velvetbean
growing wild in their fields and noted its
ability to smother weeds and improve maize
yields. They collected seed and broadcast it
over a larger area, giving rise to a practice
known as "making a fallow field" (hacer acaual).
According to experienced farmers, weeds are
eliminated by the aggressive velvetbean crop,
and soil fertility is regained. In the neighboring
ejido of Mecayapan, farmers plant velvetbean at
the end of the dry season in winter maize fields
and allow it to develop as a sole crop
throughout the summer season. The abundant
velvetbean growth, known locally as a picapical,
is slashed in November, and winter maize is
planted into the mat of decomposing leaves
and vines, where it develops relatively free of
competition from weeds.

On-farm research with velvetbean during the
early 1990s resulted in the development of
more intensive management strategies in
summer and winter maize. These strategies
have the potential to improve maize
productivity and slow the decline of soil
fertility (Buckles and Perales 1995). The
research determined that farmers are willing to
plant mucuna into summer maize 40-60 days
after the maize with a view to improving soil
fertility and controlling weeds. While no long-
term data on the soil fertility implications of
this practice have been collected for the Sierra
de Santa Marta, evidence from similar locations
suggests that fertilizer substitution rates in










maize of 60-80 kg N/ha following a mucuna
relay crop can be attained (Lobo Burle et al.
1992; Moscoso and Raun 1991). In addition, the
effects of weed control are probably significant.
Qualitative observations from the Sierra de
Santa Marta suggest that weed populations are
reduced by two thirds in fields managed with
mucuna relay crops compared to current
farmer management (Buckles and Perales
1995). These effects are undoubtedly quite
variable, however, and rely entirely on the
conservation of the residues left by the cover
crop, an issue discussed below in relation to
regional burning practices.

The impact on erosion of planting a mucuna
relay crop in summer maize has not been
determined for the Sierra de Santa Marta, but
experiments under similar conditions in
Chiapas suggest that soil loss is reduced from
50 t/ha/yr under traditional management to
4 t/ha/yr with a mucuna relay crop (L6pez
1993, p. 86). The impact of reduced erosion on
the productivity of the maize cropping system
is unknown, but given that soils in the Sierra de
Santa Marta are a limited resource highly
susceptible to erosion, this benefit can be
assumed to contribute to the long-term
sustainability of maize production.

The direct costs associated with mucuna relay
crops are minimal. Fields are weeded normally
before planting the cover crop, and 2-4 seeds
are planted every meter or so using the maize
rows as guideposts. Depending upon the
timing of mucuna planting, maize harvesting
may be encumbered by the abundant growth of
vines, and volunteer plants may need to be
controlled to avoid competition with maize.
An aggressive cover crop such as mucuna is
incompatible with food intercrops and
consequently imposes a significant opportunity
cost in areas where beans and other intercrops
are planted in association with maize.


Nevertheless, in monocropped maize systems
now common in the Sierra de Santa Marta,
these indirect costs are not encountered.

Mucuna is not the only legume suitable for use
as a cover crop in the Sierra de Santa Marta.
Canavalia ensiformis is also well adapted to the
regional agroecology and has been used
successfully as a cover crop by regional
farmers. The management of this crop is similar
to that of mucuna, although it can be
intercropped in summer maize earlier, thereby
providing additional erosion-control benefits.
Similar fertilizer substitution rates and weed
control effects have also been reported for
canavalia (Barreto 1994). Other legumes such as
"arnica" (Tithonia sp.), pigeon pea (Cajanus
cajan) and "chipilfn" (Crotalaria sp.), all of which
are known in the region, have also been used
by farmers as cover crops or improved fallows
with positive results. There is, however, a need
to evaluate other species of legumes as cover
crops, especially for use in the cooler climate of
the upland zone.

While the cover crop management practices
outlined above can help intensify maize
systems and improve their productivity,
various factors affect their potential impact and
adoptability (Buckles and Barreto 1996; Buckles
and Perales 1995; Soule 1995). First, the
magnitude of cover crop impacts on maize
productivity varies considerably from field to
field, especially during the first few years of
use. This is largely the result of variations in
the quantity of biomass produced by the crop
during a growing season. Empirical evidence
from various sites suggests that a minimum of
2-3 t/ha of dry matter is needed for cover crops
to significantly improve maize productivity
(Buckles and Barreto 1995; Lobo Burle et al.
1992). In the Sierra de Santa Marta, mucuna
does not grow vigorously in the first year on all
fields; more commonly mucuna growth does










not reach this minimum during the first year or
two. Qualitative evaluation suggests, however,
that even these relatively poor fields can
produce more than 3 t/ha of mucuna dry
matter after two or three years of continuous
management as a relay crop in summer maize.
Initial nodulation problems, the degraded
condition of many maize fields in the Sierra de
Santa Marta and the positive incremental
impacts on soil conditions of continuous
mucuna use may explain this initial variability
and improvement over time. One implication is
that two to three years of continuous mucuna
relay cropping in summer maize may be
needed before significant benefits are realized
on most fields. This feature of the technology,
in turn, has implications for farmer planning
horizons and the use of incentives, issues
discussed below.

Second, the delayed impact of cover crops on
maize productivity means that only farmers
with secure access to land can be expected to
adopt the technology. Uncertain access to land
in ejidos that are dividing collective lands into
individual parcels has proven to be a significant
barrier to the adoption of cover crops and
hedgerow practices in these communities.

Third, farmers must have access to seed and
accurate information on the use of cover crops.
Preliminary extension efforts with velvetbean
in the Sierra described below indicate that local
knowledge of this practice can be accelerated
by supporting farmer-to-farmer communication
and adaptive research (Buckles, Pare, and
Arteaga 1995).

Commercial fertilizers
Empirical evidence to date suggests that while
cover crops can improve soil fertility and


reduce weed populations, they cannot on their
own solve problems of low soil productivity. The
use of plant nutrients available or generated on
the farm by cover crops in tandem with
moderate amounts of externally derived
nutrients (e.g., commercial fertilizer) may be the
most appropriate approach to soil productivity
constraints when fertilizers are costly and the
impact of cover crops relatively slow.

While experimental evidence of maize response
to commercial fertilizers in the Sierra de Santa
Marta is limited, most maize fields would
probably benefit from the application of
commercial fertilizers at rates higher than those
currently used. Improved methods of fertilizer
application may also improve the efficiency of
fertilizer use. Empirical evidence suggests that
the application of 75 kg/ha of di-ammonium
phosphate (18-46-00) at maize planting, and 100
kg/ha of urea (46-00-00) 30 days after planting,
both buried at the base of the maize plant, could
improve maize yields over non-fertilized fields
by 700 kg/ha, (i.e., over 50% of average maize
yields) (see Buckles and Perales 1995; Bello 1994;
Tasistro 1994). This is a dramatic response to a
single input and a clear indicator of the degree to
which low soil fertility constrains maize
productivity. In the upland zone, where fertilizer
use is still virtually absent, the above rates seem
adequate. However, in the relatively more
degraded lowland zone, with a longer (though
still recent) use of fertilizer, higher rates may be
needed to give an initial boost to productivity.
Although fertilizer application efficiency can be
raised by burying the fertilizer,6 this is a rather
laborious process and may not be directly
adopted by farmers in the region. A lower
response to fertilizer than the experimentally
measured response thus seems likely.


6 Especially some of the N-fertilizers such as urea and ammonium sulfate, which can lose substantial amounts of
nutrient through NH3 volatilization.










The relatively high initial level of commercial
fertilizer needed to boost productivity may be
reduced after several years of cover crop use
and the application of other soil conservation
practices. However, some fertilizer input
(especially P) may remain a necessity for some
time to ameliorate severe soil fertility problems.
Land degradation in the Sierra de Santa Marta
has progressed to such a degree that recovery
of historic levels of soil fertility may not be
possible.

Conservation of crop residues
The conservation of crop residues as "dead
mulch"7 in the field has a number of
advantages, including soil conservation, water
conservation, and weed control. A soil
conservation effect is achieved through the
presence of the mulch as a protective layer that
reduces the erosive impact of the rain. In
addition, the mulch reduces erosion by slowing
runoff with new physical barriers and by
improving the soil's physical structure (and
thereby increasing water infiltration). Both
these features are of special relevance in a
region where tropical rain storms are common
and steep slopes are cultivated.8 The water
conservation effect is mainly a result of
reduced water losses (less runoff, more
infiltration, less evaporation) resulting in more
available soil moisture, which can reduce the
productive losses during dry spells, especially
during the tapachole season. The weed-control
effect is mainly a result of inhibiting weed
emergence.

To achieve these beneficial effects, farmers
must maintain sufficient residue9 to form a


mulch layer. In the study area, such benefits are
often undermined by the burning that occurs
during land preparation for the temporal
season. This practice tends to reduce all residue
to ashes and to leave fields entirely bare and
extremely susceptible to erosion. The practice
also eliminates the residues of cover crops,
annulling their benefits as well. To realize the
benefits of conserving crop residues, farmers
should also limit soil movement during
weeding, especially on steeper fields. To
achieve an adequate weed-control effect
without substantially increasing labor costs,
farmers can make limited use of herbicides.

Again, no long-term data on the impact of this
practice have been collected for the Sierra de
Santa Marta, but evidence from experiments
under similar conditions elsewhere suggests
that the impact can be substantial. On a 65%
slope in a similar region in Chiapas, for
example, soil erosion was reduced from 50 t/
ha/yr under traditional management to 4.6 t/
ha/yr with conservation of crop residues
(L6pez 1993, p. 92). On a 15% slope in a similar
region in Nigeria, soil erosion was reduced
from 13.4 t/ha/yr under traditional
management to 0.1 t/ha/yr with conservation
of crop residues (Lal 1976, as cited by Tasistro
1994, p. 90). Reduced erosion and other benefits
translate into substantially higher yields in the
mid- to long term. Data from other similar
conditions in Nigeria also suggest a 1 ton yield
differential (amounting to a 30% yield increase)
in favor of conserving crop residues from the
third to fifth year of continuous use (Ezumah
1983, as cited by Tasistro 1994, p. 96).


7 In contrast to the term "living mulch" used to describe cover crops.
8 The soil conservation effect of the dead mulch is generally complementary to that of the living mulch, as the dead
mulch is most effective at the on-set of the rainy season when the living mulch is still absent or still in the process of
establishment.
9 Of crops, weeds, or other vegetative material.










In the Sierra de Santa Marta, direct costs
associated with residue conservation are
mainly related to the use of herbicides, which
present direct cash costs,10 an important
consideration in a subsistence-oriented
agricultural system. However, herbicides
allow for major labor savings so that the actual
costs are far from prohibitive, as is illustrated
by their already widespread adoption.
Another cost is the slightly increased labor
demand for sowing because the residues make
manual sowing a little more cumbersome.

A potentially prohibitive cost is related to
alternative uses of the crop residue, especially
as animal fodder. However, in the study area,
alternative (and better) fodder sources are
abundant, and the use of maize stover as a
fodder source is limited. Consequently, crop-
livestock interactions do not represent a
significant constraint on the conservation of
crop residues.

While the conservation of residues can help
intensify maize systems and improve their
productivity, various factors lower their
potential impact and adoptability. A major
issue is the reliance on herbicides to achieve
adequate weed control. Although herbicides
are already relatively widespread in the
lowland zone, their use in the upland zone is
still relatively limited. Furthermore, although
some farmers may have used herbicides for the
last few years, knowledge of the basic
properties of the products and application
requirements is generally limited (Tasistro
1994). Farmers in the region still need training,
especially on health and safety issues


(including issues related to environmental
safety, such as the cleaning of back-pack
sprayers in streams). Such training will increase
the cost of widespread and safe adoption of the
technology.

Another major issue concerns burning as a field-
preparation measure for the temporal
season.11 Farmers burn for a variety of reasons,
and although not all of these are applicable in
the current systems (see Tasistro 1994), the
practice is deep-rooted and difficult, albeit not
impossible, to change. An additional problem is
that, even when farmers are motivated to
conserve crop residues, others may accidentally
burn those residues. Thus, unless and until
everyone adopts no-burning practices, the
danger remains that fires will accidentally
spread from other fields or even other
regions.12 On the positive side, however, the
conservation of crop residues will potentially be
facilitated by the adoption of other
complementary activities, such as the use of
cover crops and hedgerows.

Contoured hedgerows and maize rows
While the conservation of crop residues is an
effective means of reducing soil erosion and
enhancing nutrient cycling, the amount of crop
residues left on the field may provide
insufficient cover under conditions of low maize
productivity. An additional soil conservation
technique adapted to field conditions in the
Sierra de Santa Marta is the use of leguminous
trees and grasses as contoured living hedgerows
and the contour planting of maize crops. Over
time, these practices facilitate the gradual
formation of terraces.


10 Both of a recurrent (herbicides) as well as an investment (back-pack sprayer) nature.
11 Interestingly enough, all farmers who grow maize in the tapachole season already conserve their crop residues, albeit
frequently as "standing" mulch. This is, however, mainly related to the necessary overlap between the two seasons:
the temporal crop cannot yet be harvested when the tapachole crop needs to be established.
12 Occasionally fires travel large distances, especially towards the end of the dry season. Burning is a relatively common
practice in livestock areas to regenerate pasture, and in 1994, for example, pasture fires in the lowlands reached up to
the Soteapan ejido in the lowland zone.










Contoured hedgerows, an agroforestry practice
known in various parts of the world, were
adapted to the neighboring Sierra de Los
Tuxtlas, Veracruz, by Mexico's national
agricultural research system (INIFAP) (Z6fiiga
et al. 1993). The technology involves the
establishment of contoured hedgerows using
Gliricidia sepium, a leguminous tree native to
the region. Contour guidelines are traced on
the field at distances determined by the slope
of the land (on steeper land, the contour lines
are closer together; on flatter land, they are
farther apart). A small furrow is made on the
contoured line into which gliricidia seeds are
placed and covered. Once the crop germinates,
phosphate fertilizer is applied to the hedgerow,
and it is weeded once or twice during the first
season. In subsequent seasons, crop residues
are used to reinforce the base of the hedgerow
while the hedgerow itself is pruned twice a
year to 30-40 cm so that it does not shade field
crops. Replanting to fill gaps in the hedgerows
may also be required in the second year. In all,
some 22 person days per hectare are required
to establish the hedgerows, and 5 person days
per hectare are needed for annual maintenance
(Z6fiiga et al. 1993).

The main conservation benefit of contoured
hedgerows is their potential to reduce soil
erosion. The velocity of rain runoff down field
slopes is greatly diminished by contoured
hedges at periodic intervals, an effect that in
turn reduces the extent of sheet erosion. Under
experimental conditions, soil losses have been
reduced with hedgerows compared to farmer
practices in the Sierra de Los Tuxtlas (Oropez,
Rios, and Nicolas 1994).

Another potential benefit associated with the
establishment of contoured hedgerows is the


potential modification of farmer land-
preparation and maize-planting practices. As
noted above, farmers in the Sierra de Santa
Marta burn crop residues prior to planting
maize in rows up and down the slope of fields,
practices that under current circumstances
increase soil erosion. Nevertheless, while some
farmers with contoured hedgerows already
avoid burning crop residues (because burning
may damage the hedgerow) and plant their
maize in contours by using the hedgerow as a
guide, most farmers experimenting with
hedgerows have not modified these features of
their farming practice. Further farmer
experience will reveal whether adoption of
these additional practices can be expected.

While gliricidia has proven to be an effective
material for the establishment of hedgerows, the
plant provides no direct economic benefit. As a
result, adaptive research with other hedgerow
materials has been initiated in the region by the
PSSM to identify potential dual-purpose
hedgerow species (i.e., soil conservation and
crop production) that could significantly
increase the profitability of the practice.13

Recent extension efforts with contoured
hedgerows using gliricidia in the Sierra de Santa
Marta suggest that the technology is of interest
to regional farmers (Buckles, Pare, and Arteaga
1995). However, various factors affect the
potential of the technology. First, farmers must
have access to gliricidia seed and the technical
assistance needed to establish the crop (notably
the contours). Although the seed is available
locally, it must be collected and transported in
relatively large quantities. Farmer promoters
trained by the PSSM demonstrated their ability
to provide technical assistance during the 1994
summer season when contoured hedgerows


13 The PSSM is currently assessing lemon grass (Cymbopogon citratus) for use as a hedgerow material; lemon grass is
currently grown on house compounds in the region for use as tea. The stem of the grass is an ingredient in Thai
cuisine popular in the United States, where commercial opportunities are being explored by the PSSM.










were established on more than 300 hectares of
farmland in the context of an inter-institutional
extension campaign described below.

Second, the erosion-control benefits of
hedgerows are not immediate. Preliminary
evidence from the Sierra de Los Tuxtlas
suggests that the erosion-control benefits of the
hedgerows are generally not realized until after
two years. This gap between short-term costs
and longer-term benefits is an important feature
of contoured hedgerows, with implications for
farmer planning horizons and estimates of the
costs of adoption presented below.

Increased maize planting density
As noted, low maize planting densities
currently limit the productivity per unit of land
for maize-based cropping systems in the Sierra
de Santa Marta. Alternative maize planting
densities cannot be determined at this time,
however, in the absence of further adaptive
research on the interaction of planting density
with maize variety and varying levels of soil
fertility. Increased plant density would require
maize varieties of an appropriate height so as to
minimize competition for light under higher
densities. Also, the potential of higher plant
densities would depend upon the medium-term
impacts on soil fertility of other technologies
such as cover crops and contoured hedgerows.

Other considerations
The technological options described above
apply broadly speaking to both the lowland and
the upland zones. Environmental factors may
constrain the performance of both cover crops
and contoured hedgerows in the upland zone,
at least with the proposed plant species. Cooler
temperatures in the upland zone seem to slow
the growth of mucuna, which attained about


two thirds of the biomass achieved in the
lowland zone.14 Gliricidia hedgerows were also
about two thirds as high in the upland zone
after one season of growth as compared to
hedgerows in the lowland zone. For these
reasons, impacts of the two technologies in the
upland zone can be expected to be delayed
further than in the lowland case, and impacts
may be less pronounced over the long term. The
following economic analysis takes these
differences into account. If production
constraints continue, the cover crop and
contoured hedgerow systems currently
proposed would need to be modified through
research with other plant species and possibly
management practices as well.

The Economics of Intensification

The previous section outlined alternative
strategies for intensifying maize-based cropping
systems with special reference to their
productivity and sustainability impacts in
qualitative terms. This section will elaborate on
these impacts in economic terms. First, however,
a few observations are in order.

The present study is not based on a
comprehensive survey of the area. Therefore,
the various estimates of input and output levels
as well as farm resource endowments are only
indicative. Nonetheless, these values are
estimated with reference to extensive field work
in the area over several years and consequently
can be considered reasonably precise and
reliable for the purpose of this study.

The subsequent analysis examines the economic
effects of intensification on a model farm in each
of the two zones. The term model is used here in
the sense of typical and representative and


14 In some very high communities (e.g. the ejido Santa Marta), mucuna did not grow well at all.










consequently, as with any typology of reality,
some farms in each zone will approximate this
typology better than others. Furthermore, the
typology emphasizes some of the differences
between the two zones both in terms of
technology use (e.g., fertilizer and herbicide
use) and resource endowments.

In conventional agricultural economics,
benefits and costs are usually expressed on an
area basis. However, in extensive land use
systems, land is not the most limiting factor. In
fact, the forces that made such land use systems
extensive are generally directly related to the
scarcity of other production factors such as
capital and labor.15 Consequently, looking at
intensification solely on an area basis is
inadequate, and a systems perspective that
takes into account the various factors of
production seems more appropriate. In the
following, we will therefore first assess the
implications of intensification at the field level
(i.e., for a specific unit area in the model farm
in the with and without situation). The field-
level data are subsequently used as input to
assess the farm-level implications (i.e., for the
entire model farm in the with and without
situation).

In addition, the entire analysis is based on a
number of assumptions, the most important of
which are that the analysis:

* Limits itself to the farm level. As a result,
financial on-farm prices are used in the
calculations, and off-farm effects
(externalities) are excluded.


* Assumes that maize production remains
subsistence oriented over time. It is assumed
that the main objective of maize production
is to safeguard subsistence needs. Area and
input use on the model farm are geared to
achieve this objective, not to produce a
marketable surplus. As a result,
intensification can be expected to decrease
the maize area required to meet these needs.
This response to intensification reflects
factors that dissuade farmers from
producing a marketable surplus, factors such
as the limited (even negative) financial
returns to maize cultivation (see below),
weak marketing channels, and restricted
availability of labor and capital.

* Implicitly assumes a negligible opportunity cost
of land. In a remote extensive land use
system, the opportunity cost of land is very
low. Consequently the analysis does not
deduct the cost of land from the various
economic indicators, but rather calculates the
returns to the specific factors of production,
including land. Furthermore, the current
analysis attaches no specific value to the land
freed by intensification of the maize
production system. Intensification
theoretically frees land for more maize area,
diversification into other agricultural land
uses, or conversion to secondary forest. The
economic benefits derived from these
alternative uses have not been included in
this analysis, but would make the with case
potentially more attractive.16


15 It can be argued that the breakdown of extensive slash-and-burn systems is directly related to a reduction in the
availability of land, an effect that would seem to suggest that land is the most limiting factor. However, such an argument
ignores the simultaneous constraints imposed by the other factors of production, or implicitly assumes that the levels of
these other factors are fixed. We argue that it is within the limitations imposed by labor and capital constraints that land
may eventually become limiting in extensive systems. The breakdown arises from the reduction of fallow periods and the
lengthening of cultivated periods without an adequate adjustment of the existing ("traditional") practices to the new
circumstances, adjustments that would simultaneously ease the land constraint, at least in the near term.










* Assumes a 15% discount rate. All cost and
benefits are discounted17 at this rate to bring
them to their present value. It is assumed
that this rate adequately reflects the time
preference of the farmer. However, as this
rate is difficult to estimate empirically, the
analysis also includes the sensitivity of the
outcomes to changes in the discount rate.

The analysis follows the accounting convention
proposed by Gittinger (1982, p. 95) (i.e., that
each transaction occurs at the end of each year).
In addition, this convention reserves Year 1 for
initial investments. As a result, Year 2 is the
first accounting period in which increases in
operating cost and incremental benefits
occur.18 Due to the differences between the
upland and lowland zone, results are presented
separately for each.



Upland Zone: Field-level implications
Cropping pattern. It is assumed that in the
actual situation (the without case) the model
farmer cultivates a plot for three consecutive
years. Afterwards, the respective plot is
fallowed for six years before being cleared and
cultivated again. As a result, the entire cycle
amounts to nine years, resulting in a land use
intensity (LUI) of only 33%.

The previous section has already presented a
combination of technologies which would
allow for a sustainable intensification of the
current system in the upland zone. It is
assumed that under this combination


(the with case), fallow periods can be drastically
reduced to only one year due to the combined
use of fertility increasing technologies (cover
crops and commercial fertilizer) as well as a
better conservation of the available fertility
(cover crops, conservation tillage, and
contoured hedgerows). If we assume that the
cultivation period will remain three years, the
entire cycle amounts to four years, resulting in
a land use intensity of 75%. This apparently
drastic intensification is also facilitated by the
fact that the upland zone allows for only one
maize crop a year (i.e., the fields are actually
resting for more than half of the time even
during the three years of cultivation).19 This
inter-season fallow allows fields to regenerate
to a certain degree, especially in the with case
under a cover crop.

Crop budget. The without case crop
budget20 confirms the low profitability of the
current maize production systems. In the
absence of external inputs, yields are low and
decline rapidly over the three-year cultivation
period, from a high of 1.25 ton per ha to a low
of 0.8. Labor use per unit area is relatively
constant, amounting to 76-80 labor days per ha
cultivated. The relatively constant labor needs
are the result of two forces that largely offset
each other over time: weed problems increase
the demand for weeding; declining yields
decrease the demand for harvesting and
processing. However, although labor needs
remain relatively constant, labor productivity
declines rapidly over the cultivation period:
from a high of nearly 16 kg of maize per labor


16 The potential benefits derived from agricultural diversification will be examined in another study in the collaborative
GEF/PSSM/CIMMYT effort to estimate the incremental cost of conservation. Including them here would result in
double-counting for the overall study.
17 Discounting is used to take into account the differences in terms of timing between the cost and the benefits streams.
18 As a result, the first year is the same for the with and without case, with the exception of initial investments in the with
case.
19 Although this interseason fallow includes a pronounced dry period (March-April), soil moisture in the other months
is sufficient to allow for continued weed growth and biological activity.
20 Tables A-2 and A-3 present detailed crop budgets for the without and with case in the upland zone.










day in the first year of cultivation to a low of
about 10 kg per labor day in the third and last
year of use. Valuing labor at its opportunity
cost (assumed equal to the local wage rate of a
day laborer) generates a negative net benefit
(or return to land) in each year. As a result, the
return to a labor day (and land) does not
surpass the assumed opportunity cost of US$
2.5, and furthermore drops from US$ 2.4 to 1.3
per labor day over the three years. Maize
production in the current system is relatively
uncompetitive with imported maize, as
production costs amount to US$ 170 per ton in
the first year of cultivation and spiral up to US$
250 per ton in the third. However, when one
considers farmers' subsistence orientation and
limited alternative sources of income, maize
production remains of vital importance to farm
households.

The with case crop budget provides a more
favorable outlook, although overall
profitability remains low. With the
combination of limited external inputs and
vegetative conservation measures, yields are
raised and eventually maintained at about 2.2
tons per ha. Labor use per unit area is increased
substantially, especially in the first year of
cultivation (Year 2), but eventually oscillates
around 100-105 labor days per ha cultivated.
As a result, labor productivity increases from
an initial low of 14 kg per labor day to stabilize
around 21-22 kg per labor day from the fourth
year. Valuing labor at its opportunity cost
generates a positive net benefit (or return to
land) only after the sixth year. As a result, the
return to labor day (and land) hovers below the
assumed opportunity cost of US$ 2.5 in the first
five years (the first and fifth year are fallow),
with a low of US$ 1.2 per day in the second
year and climbing to about US$ 2.5 per day in
the fourth. In the sixth and subsequent non-


fallow years, the return to labor oscillates
between US$ 2.6 and 2.8 per labor day. Maize
production in the proposed system becomes
gradually more competitive with imported
maize, as production costs decrease from US$
230 per ton in the first year of cultivation to
US$ 130 per ton in the later years.

Upland Zone: Farm-level implications
The field-level implications seem to suggest
that intensification could make maize
production a little more attractive, at least in
the mid-term on a unit area basis. However, as
stated, land is relatively abundant in extensive
systems and an 'extensive' land use system
may be entirely rational in view of limited
resources (including labor) and a subsistence
orientation. To make the comparison more
realistic, we therefore need to consider the
farm-level implications of intensification.

Land use. Figure 4 conceptualizes the land-use
implications of the actual and proposed
cropping patterns for the "model" farm in the
upland zone.21 It is assumed that in the actual
situation (the without case), the farmer
cultivates a plot for three consecutive years and
subsequently shifts his productive activities to
a new plot. As a result, although the farmer
only cultivates one plot, he needs two similarly
sized plots resting at any given time. We
assume 2 ha of maize are cultivated annually
on the model farm, an amount which, on
average, barely covers subsistence needs. The
entire system would thus require 6 ha of
cultivable land per household.

It is assumed that in the improved situation
(the with case) the model farmer adopts the
new system progressively on his lands, thereby
rotating the portion left fallow over his fields.
Such a gradual implementation is preferable


21 In view of the accounting convention adopted, the first year is similar for both the with and without case.












for farmers as it would satisfy both their annual
consumption and fallowing needs, and
simultaneously smooth the labor peaks. For
various reasons, the comparison begins with a
new cropping cycle (in Year 2).22


Nevertheless, changes in the starting stage are
not expected to result in major changes in the
overall outcome of the analysis.23 Furthermore
we assume that under the improved situation
annual maize area will be reduced to 1.5 ha from
the original 2 ha per household (as production


ACTUAL (without case)
Field
Year Season A


PROPOSED (with case)
Field
Year Season Al


Temp


Temp F
1
2
3
F,


1 Fl
2 1
3 2
FI 3
1 Fi


A2 A3 A4 C


F
la F
2a la
3 2a
F, 3
1 F,
2 1
3 2


cycle


Where:
F
m--


Maize (actual)
Fallow (actual)
Cropping cycle


At equilibrium at any given time:
Plots in production
Plots fallowed
Plot area
Annual maize area
Annual system area
Duration use
Duration fallow
Cycle length

LUI Annual


Where: M Maize (improved)
n Maize (actual)
Fi Fallow (improved)


1 plot(s)
2
2 ha
2 ha
6
3 years
6
9

33.3%


F Fallow (actual)
At equilibrium at any given time:
(Sub)Plots in production
(Sub)Plots fallowed
(Sub)Plot area
Annual maize area
Annual system area
Duration use
Duration fallow
Cycle length


a Adjusted

SCropping


3 plot(s)
1
0.5 ha
1.5 ha
2
3 years
1
4


LUI Annual 75.0%
Note: Field A divided in 4 sub-plots; fields B+C (2 ha each)
available for other purposes (C after year 1). Adjusted
reflects lower yields due to reduced fallow period.


Figure 4. Observed and proposed cropping patterns for the model farm in the buffer zone,
Sierra de Santa Marta, Veracruz.


22 A major reason is simplicity. Including additional starting points complicates the analysis and presentation with limited added
precision. In addition, coefficients in the initial years of the with case are based on implementation in a rested field.
Implementation in used fields would require scaling down the relevant coefficients. Another reason is realism. For an
adequate comparison between the two systems over time, both should start at a similar reference point (first year after fallow).
This is especially true in the upland zone where yields and benefits drop substantially over the 3-year cultivation period.
23 This is mainly because a different starting stage would only affect the initial years. In subsequent years the equilibrium
situation would prevail anyway whatever the starting stage. Even so, the expected difference in the initial years will be limited
because: the effect partly cancels out during the transition years because part of the area is still under the observed system
anyway; a different starting stage would require the scaling down of technical coefficients, so that the net effect will probably
be marginal.


7
7
7
7










will still be largely subsistence oriented, whereas
productivity is higher). As still 25% of the area
would remain fallow, the improved system
would require 2 ha of cultivable land per
household, a substantial reduction from the
original 6 ha. The land freed from maize
production (4 ha per farm household in the
upland zone) could become available for
agricultural diversification and/or conversion to
secondary forest, thereby reducing the pressure
of the maize cropping system on the forest
margins.

Economic implications.24 Going from the field
to the farm level naturally does not make the
economics of maize cultivation in either the with
or without case more attractive. However,
expanding to the farm level does allow for a
more adequate assessment of the implications of
adopting the with case on the model farm and of
the relative savings or additional costs this
would imply.

To assess the economic farm-level implications,
we have estimated the annual costs and benefits
in relation to maize cultivation at the farm level
over the time period considered (equal to the
cycle length of the actual system) for both the
with and without case. These have subsequently
been discounted at a 15% rate.

Most interestingly, adoption of the with case
does not substantially alter the labor demand at
the farm level. In fact, over the period
considered, the with case would generate a net
savings of some 20 days. Although labor
demand per unit area is substantially higher for
the with case, this is fully offset by the smaller
productive area. In addition, the with case
alleviates the labor bottleneck in the busy
months of May-July.25


Farm production levels are not hampered by
the reduction in cropped area, as this is offset
by the substantially higher yield levels. In fact,
the with case only runs a relative deficit
(compared to the without case) in the second
year, whereas in the subsequent years the gap
widens in favor of the with case to eventually
oscillate between an additional 750 and 1,700
kg annually. Over the entire period, the with
case produces an additional 8 tons of maize per
farm, averaging 800 kg annually. This relative
surplus would ensure that subsistence needs
are more adequately met (under the current
system households occasionally do not meet
subsistence needs). In addition, some of the
increased production could be sold locally.

Although farm-level labor requirements are
similar in the with and without case, physical
input requirements are substantially larger in
the with case. As a result, total input costs are
higher in the with case, amounting to a
discounted increase of US$ 190 per farm over
the period considered. But total production is
also larger in the with case, resulting in a
discounted increase in gross benefits of US$ 360
per farm, or a benefit-cost (B/C) ratio of 1.9.
Deducting the increased costs from the
increased gross benefits would generate a net
benefit of US$ 170 per farm over the time
period considered.

Notwithstanding the fact that the with case
seems to be economically more viable than the
without case over the entire time period
considered, there is a marked difference
between the early and late years. In particular,
the first five years of the with case are not
especially attractive. In fact, changing from the
without to the with case would imply a net
deficit of approximately US$ 70 over these five


24 Table C-1 presents the detailed outcome of the cost-benefit analysis (CBA) for the with and without situation in the
upland zone. Here only the salient results will be highlighted.
25 Appendix B presents the temporal distribution of labor (by month and year) in more detail.










years. This is mainly a result of the substantially
larger input requirements (a discounted
increase of US$ 140) which is only partially
offset by higher gross benefits (a discounted
increase of US$ 70), resulting in a B/C ratio of
0.5. Once fully established, the with case is
substantially more attractive than the without
case, generating a net surplus of US$ 230 over
the subsequent five years, resulting in a B/C
ratio of 5.6. However, some form of incentive
may well be required to overcome the
prohibitively expensive initial period.

The annual B/C ratios further support the need
for some incentive in the initial period. Because
resource-poor farmers in the tropics operate in
high-risk conditions, it is generally accepted
that the B/C ratios of new technologies should
surpass 2.0 to be attractive to farmers
(CIMMYT, 1988, pp. 34-35). In none of the first
five years does the B/C ratio surpass this
threshold value, but in the subsequent years it
easily does so. This seems to suggest that the
incremental costs of adopting the with case
generate sufficient benefits after the sixth year
to be self-supporting (i.e., any calculation of the
"incremental cost" of adoption for farmers need
only consider the first five years).
In view of the above, it should be clear that
simply alleviating the US$ 70 deficit in the first
five years would not be enough to lure farmers
into adopting the with case, as this would only
raise the B/C ratio to a meager 1.0 during the
start-up period. It may be more appropriate to
cover all the incremental costs farmers incur
during the start-up period (i.e., a discounted
sum of US$ 140 per farm household). The


additional benefits generated during the start-
up phase would then provide a further
incentive to the farmer to adopt the with case.26

Sensitivity analysis.27 The following section
briefly presents the sensitivity of the farm-level
implications to changes in the discount rate, the
opportunity cost of labor and the assumed
yield increase in the with case.

* Discount rate. In view of the timing of costs
and benefits, it is not surprising that a lower
discount rate favors the with situation.
Lowering the discount rate to 10% raises the
gross benefits more than the input costs,
resulting in an overall 54% higher net
benefit. Conversely, increasing the discount
rate to 20% would lower the gross benefits
more than the input costs, resulting in a 35%
lower net benefit. The incremental costs
during the start-up phase amount to a
discounted US$ 165 and 115 per farm
household, for the 10 and 20% discount rates
respectively. Overall, however, most of the
observed differences between the with and
without case remain valid and are not very
sensitive to changes in the discount rate.

* Opportunity cost of labor. Although labor is
the major input in the production process in
the with and without cases, the solution is not
very sensitive to the actual value of the
opportunity cost of labor. This is not
surprising: recall that the above discussion
showed that labor demand at the farm level
is largely similar for the with and without
situation.


26 Strictly speaking, a lesser amount of cost sharing may suffice to raise the B/C ratio above the threshold value.
However, in view of compatibility with the other study components we use the sum of incremental costs to farmers
during the start-up period, without correcting for increased benefits in this phase, along the lines suggested by
Cervigni (1995, p. 8).
27 Table C-2 presents the detailed outcome of the sensitivity analysis in the upland zone. Here, only the salient results
will be highlighted.










* Assumed yield increase. The original with case
assumes an initial yield increase of 40%
relative to the without case.28 Reducing the
initial yield increase to only 30% would
naturally make the with case less attractive
because the annual gross benefit is reduced
proportionally. However, lower yields
would also reduce labor needs for harvesting
and shelling (resulting in a 70-day decrease
relative to the without case over the entire 10-
year time period), so that labor costs would
also be reduced. As a result, the gross benefit
of changing will drop to US$ 220 per farm
whereas the total costs will drop to US$ 140.
This results in a net benefit of only US$ 80
per farm over the time period considered (a
54% decrease) and a B/C ratio of only 1.6.
On average, the model farm would produce
an additional 560 kg of maize annually in the
with case, still sufficient to more adequately
meet subsistence needs. The initial five years,
however, would become even less attractive,
presenting a cumulative discounted net loss
of US$ 100 and a B/C ratio of 0.1. This
decrease is mainly the result of a substantial
decrease in benefits, as the incremental costs
in the start-up period only drop to US$ 120
per farm household.

Increasing the initial yield increase to 50%
would naturally make the with case more
attractive, although labor demand would
also be affected (resulting in a 30-day
increase relative to the without case). Overall,
the net benefit would be increased to an
accumulated discounted total of US$ 260
over the 10-year time period (a 54%
increase), resulting in a B/C ratio of 2.1. The
start-up period will also become relatively
more attractive, generating a net loss of only


US$ 30. The reduced loss is mainly the effect
of a substantial increase in benefits, as the
incremental costs in the start-up period
increase to US$ 160 per farm household.

Lowland zone: Field-level implications
Cropping pattern. It is assumed that in the
actual situation (the without case) the model
farmer cultivates a plot for three consecutive
seasons (1.5 years, including two temporal and
one tapachole season). Afterwards, the
respective plot is fallowed for five seasons (2.5
years, including two temporal and three
tapachole seasons) before being cleared and
cultivated again. As a result the entire cycle
amounts to four years, resulting in an annual
land use intensity (LUI) of 75% (or a seasonal
LUI of only 38%). 29 This relatively high level of
cropping intensity is unsustainable under the
current cropping practices.

It is assumed that in the with case the number
of fallow seasons can be reduced to three
(including one temporal and two tapachole
seasons, totaling 1.5 years of fallow) for each
four years due to the combined use of fertility-
increasing technologies (cover crops and
fertilizer) as well as a better conservation of the
available fertility (cover crops, conservation
tillage, and contoured hedgerows). For the
same cycle length, the number of crops
increases to five (including three temporal and
two tapachole seasons, totaling 2.5 years of
cultivation). As a result, annual land use
intensity would increase to 125% (or seasonal
LUI to 63%).

Crop budget. The without case crop budget30
again confirms the low profitability of the
current maize production systems. With


28 The yield increases in the subsequent two years are a multiple of the initial increase.
29 Annual LUI is calculated here as the number of crops divided by the number of years; seasonal LUI is calculated as
the number of crops divided by the number of seasons.
30 Table A-3 and A-4 present the detailed crop budgets for the without and with case in the lowland zone.










limited use of external inputs in the lowland
zone, yields are low and decline rapidly over
the three-season cultivation period. Temporal
season yields decline from a high of 1.4 ton per
ha in the first year of cultivation to a low of 1.0
ton per ha in the second year, whereas tapachole
yields average only 0.5 ton per ha. Labor use
per unit area is relatively constant in the
temporal season, amounting to around 75 labor
days per ha cultivated, but averages less than
30 days in the tapachole season. Labor
productivity declines from 18 to 14 kg per labor
day over the cultivation period. Valuing labor
at its opportunity cost generates a negative net
benefit (or return to land) in each season. As a
result, the return to a labor day (and land) does
not surpass the assumed opportunity cost of
US$ 2.5, and drops from US$ 1.8 to 1.2 per
labor day over the cultivation period. Again,
maize production in the current production
system is relatively uncompetitive with
imported maize. Production costs increase
from US$ 180 per ton in the first season to US$
240 per ton in the last. However, as in the
upland zone, maize production remains of vital
importance to smallholder households in view
of their subsistence orientation and limited
alternative sources of income.

The with case crop budget again provides a
better outlook, although overall profitability
remains low. With a combination of increased
external input use and vegetative conservation
measures, yields are raised and eventually
maintained at about 2.5 tons per ha for the
temporal season and about 0.9 tons per ha for
the tapachole season. Labor use per unit area is
increased substantially, especially in the first
year of cultivation, but eventually oscillates
around 105 labor days per ha cultivated in the
temporal season and 30-46 days per ha
cultivated in the tapachole season. As a result,
labor productivity in the temporal season


increases from an initial low of 17 kg per labor
day to stabilize around 24 kg per labor day as
of the fourth year. Labor productivity in the
tapachole season is similar and oscillates
between 20 and 26 kg per labor day. Valuing
labor at its opportunity cost generates a
positive net benefit (or return to land) as of the
fourth year. As a result, the return to labor day
(and land) hovers below the assumed
opportunity cost of US$ 2.5 in the first three
years, with a low of US$ 1.1 per day in the
second year and climbing to about US$ 2.2 per
day in the third (the first year is fallow). In the
fourth and subsequent non-fallow years, the
return to labor oscillates between US$ 2.7 and
2.8 per labor day in the temporal season and
between US$ 2.5 and 3.3 per labor day in the
tapachole season. Maize production in the
proposed system becomes gradually more
competitive with imported maize. Production
costs decrease from US$ 220 per ton in the first
year of cultivation to stabilize between US$ 110
and 140 per ton in the later years.

Lowland zone: Farm-level implications
Land use. Figure 5 conceptualizes the land-use
implications of the actual and proposed
cropping patterns for the model farm in the
lowland zone. It is assumed that in the actual
situation (the without case) the model farmer
cultivates a plot for three consecutive seasons
(two temporal and one tapachole season). During
the second temporal season, the farmer clears a
second field which will be used for the
subsequent three seasons, abandoning the first
plot after completion of the second temporal
season. Something similar happens every year,
so that the farmer cultivates two plots each
year, progressively shifting his productive
activities to a new field each year. As a result,
the farmer will have two temporal plots and
only one tapachole plot in production each year.
If we recall that the fallow period is assumed to











total five seasons, then the farmer can return
the first plot in the fifth year of the cycle.
Therefore, whereas the farmer may only
cultivate two plots annually, he needs two
similarly sized plots resting at any given til
We assume that 1.33 ha of temporal maize a

ACTUAL (without case)
Field
Year Season A B C
1 Temp F F
TaD F F F


2 Temp
Tap
3 Temp
Tap F
4 Temp F
Tap F
5 Temp F
Tap F
6 Temp
Tap
7 Temp
Tap F
8 Temp F
Tap F
9 Temp F
Tap F
10 Temp
Tap


F F
F F

F
- F

F
F
F F
F F

F F
1


*n to




me.
nd


0.67 ha of tapachole maize are cultivated
annually on the model farm. On average, these
amounts barely cover subsistence needs. The
entire system would thus require 2.67 ha of
cultivable land per household.



PROPOSED (with case)
Field


D Year Season B2 Al A2 B1
1 Temp F F F F
Tap F F F F
2 Temp 1 F
F Tap Fi F
F 3 Temp 2 la
F Tap Fi Fi F
F 4 Temp 3 2a la
F Tap 4 Fi Fi F
5 Temp Fi 3 2a la
Tap 5 4 Fi Fi
6 Temp 1 Fi 3 2a
F Tap Fi 5 4 Fi
F 7 Temp 2 1 Fi 3
F Tap Fi Fi 5 4
F 8 Temp 3 2 1 Fi
F Tap 4 Fi Fi 5
9 Temp Fi 3 2 1
Tap 5 4 Fi Fi


10 Temp
Tap


1 Fi 3 2
Fi 5 4 Fi


Where: I Maize (actual)
F Fallow (actual)
I | Cropping cycle


At equilibrium at any given time:
Temp.
Plots in production 2
Plots allowed 2


Plot area
Annual maize area
Annual system area

Duration use/cycle
Duration fallow/cycle
Cycle length


1.33 0.67


Overall


0.67


Where: n Maize (improved) a Adjusted
= Maize (actual)
Fi Fallow (improved) m Cropping cycle
F Fallow (actual)


At equilibrium at any given time:
Temp.


3
1


plot(s) Sub)Plots in production
(Sub)Plots allowed
ha (Sub)Plot area
ha Annual maize area
D Annual system area

seasons Duration use/cycle
S Duration fallow/cycle
D Cycle length


LUI Annual 75.0% Seasonal 37.5%


Overall


1.00 0.67


plot(s)


0.33 ha
ha


seasons


Annual 125.0% Seasonal 62.5%


Note: Fields A+B each divided in 2 sub-plots; fields C+D (0.67 ha each)
available for other purposes after year 1. Adjusted reflects lower
yields due to reduced fallow period.

Figure 5. Observed and proposed cropping patterns for the model farm in the lowland zone,
Sierra de Santa Marta, Veracruz


C D

FE










It is again assumed that in the improved
situation (the with case) the model farmer adopts
the new system progressively on his lands,
thereby gradually expanding the improved
system over his plots. Furthermore we assume
that under the improved situation annual
temporal maize area will be reduced to 1.0 ha,
whereas annual tapachole maize area remains at
0.67 ha per household (because production will
still be largely subsistence oriented, whereas
productivity is higher). Because 25% of the
temporal area (and 50% of the tapachole area)
would still remain fallow, the improved system
would require 1.33 ha of cultivable land per
household (i.e., half of the original 2.67 ha), and
1.33 ha per farm household in the lowland zone
would be freed from maize production.

Economic implications.31 To assess the
economic farm-level implications, we have
estimated the annual costs and benefits in
relation to maize cultivation at the farm level
over a 10-year time period for both the with and
without case.32 The discount rate is again 15%.

Although labor demand per unit area is
substantially higher for the with case, at the farm
level this is partially offset by the smaller
productive area in the temporal season. As a
result, adoption of the with case only increases
labor demand at the farm level by some 80 days
over the 10-year time period, averaging around
11 additional days annually once the with case
stabilizes. However, the with case does not
represent a too substantial increase in labor
requirements in the busy months of May-July.33


Farm production levels are not hampered by
the reduction in cropped area, which is offset
by the substantially higher yield levels. In fact,
the with case only runs a negligible relative
deficit (compared to the without case) in the
second year, whereas in the subsequent years
the gap widens in favor of the with case to an
additional 1,200 kg annually. Over the entire
10-year period, the with case produces an
additional 7 tons of maize per farm, averaging
680 kg annually. This relative surplus would
ensure that subsistence needs are more
adequately met, whereas some of the increased
production could potentially be sold locally.

The with case presents an increase in both labor
and non-labor costs over the without case,
resulting in a discounted increase in costs of
US$ 140 per farm over the time period
considered. Total production is however also
larger in the with case, resulting in a discounted
increase in gross benefits of US$ 360 per farm,
or a benefit-cost (B/C) ratio of 2.6. Deducting
the increased costs from the increased gross
benefits would generate a net benefit of US$
220 per farm over the time period considered.

Although the with case seems to be more
economically viable than the without case over
the entire time period considered, there is again
a marked difference between the early and late
years. In particular, the first five years of the
with case are not very attractive. In fact,
changing from the without to the with case
would imply a net loss of US$ 10 over these five
years. This is the result of the larger input
requirements (a discounted increase of US$


31 Table C-3 presents the detailed outcome of the CBA for the with and without case in the lowland zone. Here only the
salient results will be highlighted.
32 To facilitate comparison, the time period for the two zones is the same. In any case, a single four-year cycle would not
be sufficient to adequately capture changes over time as the improved system only stabilizes after the fifth year of
implementation. Therefore the time period had to be expanded to at least include two full cycles of the actual system.
33 Appendix B presents the temporal distribution of labor (by month and year) in more detail.










110) not being offset by higher gross benefits (a
discounted increase of US$ 100 only), resulting
in a B/C ratio of 0.9. Once fully established, the
with case is substantially more attractive than
the without case, generating a net surplus of
US$ 230 over the subsequent five years, with a
B/C ratio of 9.7. Again, however, some form of
incentive may well be required to overcome the
relatively costly initial period.

The annual B/C ratios further support the need
for some incentive in the initial period. In none
of the first five years does the B/C ratio surpass
the 2.0 threshold value, but in the subsequent
years it easily does so. This seems to suggest
that the incremental costs of adopting the with
case generate sufficient benefits after the sixth
year to be self-supporting. If farmers are to find
the practice attractive, it may be appropriate to
cover all the incremental costs farmers incur
during the start-up period (a discounted sum
of US$ 110 per farm household). The additional
benefits generated during the start-up phase
would then provide a further incentive to the
farmer to adopt the with case.

Sensitivity analysis.34 The following section
briefly presents the sensitivity of the farm-level
implications to changes in the discount rate, the
opportunity cost of labor, and the assumed
yield increase in the with case.

* Discount rate. In view of the timing of costs
and benefits, a lower discount rate again
favors the with situation. Lowering the
discount rate to 10% raises the gross benefits
more than the input costs, resulting in an
overall 45% higher net benefit. Conversely,
increasing the discount rate to 20% would
lower the gross benefits more than the input
costs, resulting in a 30% lower net benefit.


The incremental costs during the start-up
phase amount to a discounted US$ 130 and
90 per farm household, for the 10 and 20%
discount rates respectively. Overall, however,
most of the observed differences between
the with and without case remain valid and
are not very sensitive to changes in the
discount rate.

* Opportunity cost of labor. With a marked
difference of nearly 80 labor days between the
with and without case, the solution is sensitive
to the actual value of the opportunity cost of
labor. Lowering the opportunity cost of labor
to US$ 1.5 decreases the labor cost differential
to a discounted sum of only US$ 50, resulting
in an overall 20% higher net benefit and an
overall B/C ratio of 3.7. Conversely,
increasing the opportunity cost of labor to
US$ 3.5 increases the labor cost differential to
a discounted sum of over US$ 130, resulting
in a 20% lower net benefit, and a B / C ratio of
2.0. The incremental costs during the start-up
phase amount to a discounted US$ 90 and 130
per farm household, for the lower and higher
rates respectively.

* Assumed yield increase. The original with case
assumes an initial yield increase of 40%
relative to the without case. Reducing the
initial yield increase to only 30% would again
make the with case less attractive as the
annual gross benefit is reduced
proportionally. However, lower yields would
also reduce labor needs for harvesting and
shelling (resulting in a 30-day decrease
relative to the without case over the entire 10-
year time period), so that labor costs would
also be reduced. As a result, the gross benefit
differential will drop to US$ 230 per farm,
whereas the total costs will drop to US$ 90.


34 Table C-4 presents the detailed outcome of the sensitivity analysis in lowland zone. Here only the salient results will
be highlighted.










This results in a net benefit of only US$ 140
per farm over the time period considered (a
36% decrease), and a similar B/C ratio of 2.5.
On average, the model farm would produce
an additional 460 kg of maize annually in the
with case, still sufficient to more adequately
meet subsistence needs. The first five years,
however, would become less attractive,
presenting a cumulative discounted net loss
of US$ 40 and a B/C ratio of 0.6. This
decrease is mainly the result of a substantial
decrease in benefits, as the incremental costs
in the start-up period drop to US$ 90 per
farm household.

Raising the initial yield increase to 50%
would naturally make the with case more
attractive, although labor demand would
also be affected (resulting in a 120-day
increase relative to the without case). Overall,
the net benefit would be increased to an
accumulated discounted total of US$ 300
over the 10-year time period (a 37%
increase), resulting in a B/C ratio of 2.6. The
start-up period will also become relatively
more attractive, generating a net benefit of
US$ 20. This is mainly the effect of a
substantial increase in benefits, as the
incremental costs in the start-up period
increase to US$ 130 per farm household.

Challenges for Implementation

The farm-level analysis of costs and benefits of
technical options for intensifying maize-based
cropping systems in the Sierra de Santa Marta
indicates that while the use of plant nutrients
available or generated on the farm by cover
crops in combination with vegetative barriers
and limited external inputs would eventually
make maize production substantially more
attractive, the overall profitability of the system
would remain low. Household subsistence
requirements would, however, be assured


more adequately than under the current
circumstances and some land would be freed
from maize production for other uses. The
freeing of land is the combined result of a
reduction in actual maize area (increased
productivity of subsistence-oriented
production) in addition to a reduction in the
required fallow area, amounting to 1.33 ha per
household in the lowland zone and up to 4 ha
per household in the upland zone. This land is
potentially available for the production of more
commercially oriented crops or conversion to
secondary forest, thereby reducing the pressure
of the maize cropping system on the remaining
forest resources of the region.

A major constraint is the timing of the benefits,
as adopting the proposed technologies requires
a substantial initial investment by the farmer.
In the upland zone, for example, farmers face
an incremental cost amounting to a discounted
sum of US$ 140 per farm household over the
five-year start-up period, resulting in a B/C
ratio of only 0.5 during that initial phase. In the
lowland zone, farmers face an incremental cost
amounting to a discounted sum of US$ 110 per
farm household over the five-year start-up
period, resulting in a B/C ratio of only 0.9
during that phase. After the start-up phase, the
improved technologies are potentially self-
supporting in both zones with B/C ratios easily
surpassing the 2.0 threshold. Alleviating the
incremental cost during the start-up phase
through the appropriate use of financial
incentives therefore may be enough to facilitate
farmer adoption, assuming that a broader
strategy for ongoing adaptive research and
extension is also developed.

Experience to date in the Sierra de Santa Marta
suggests that neither the incentives nor the
research and extension effort needed to
effectively implement an incentive plan can be
expected from the current national agricultural










support services. Since the 1980s, the National
Agricultural and Forestry Research Institute
(INIFAP) has been severely weakened and the
branch of the Agricultural, Ranching and Rural
Development Secretariat (SAGAR) responsible
for technical assistance has practically been
eliminated. State withdrawal from the
provision of agricultural research and
extension has been justified by arguments that
the private sector can provide these services
more efficiently and effectively. In effect,
however, the policy has cleared the rural
landscape of agricultural institutions and
severely restricted farmer access to information
and technology needed to improve
productivity. This withdrawal is perhaps less
severe in marginal areas such as the Sierra de
Santa Marta where agricultural research and
extension have always been weak anyway.

Credit has been provided in recent years to
small-scale farmers in the Sierra de Santa Marta
through programs managed by the National
Indigenous Institute (INI) and SAGDR with
funding from SEDESOL, the mega-Secretariat
established during the early 1990s to finance
poverty alleviation programs in marginal areas.
These programs have stimulated the recent
adoption of fertilizer (and herbicide) in the
lowland zone, thereby defraying increased
input costs for some farmers. Coverage has
been limited, however, to less than 10% of the
farming population (Bello 1994), virtually all of
whom are ejidatarios with secure land tenure,
leaving the poorer and younger farmers who
are not ejidatarios virtually without access to
credit. Furthermore, recovery of the credits has
been poor and the programs have not been
replenished with new funds or expanded since
they were initiated. Despite the complexities of
the technologies promoted, none of the credit
programs provides technical assistance; the
result has been inefficient and unsafe farming
practices. PROCAMPO, a program of direct


payment to farmers intended to facilitate the
reorientation of agriculture away from less
profitable basic grain production into other
agricultural activities, has also completely
neglected the provision of technical assistance
and market studies needed to facilitate this
shift. As a result, the impact of the program in
many areas, including the Sierra de Santa
Marta, has been limited or simply reinforced
existing patterns of agricultural decline. For
example, farmers in the region have cleared
more forested land for maize production so as
to qualify for PROCAMPO subsidies, but
without increasing the productivity of the land
already under cultivation (Buckles, Pare, and
Arteaga 1995).

Over the years, there has been virtually no
adaptive research with agricultural innovations
in the Sierra de Santa Marta and no systematic
study of technology needs, two activities that
are essential for the successful implementation
of the alternative strategies outlined above.
While they appear to be robust practices (based
on limited local experience and experience in
other regions), learning costs, including further
adaptation and extension to the general
farming population, are likely to be
considerable. The virtual withdrawal of the
public sector from agricultural research and
extension, however, leaves an institutional
vacuum in the countryside; farmers in the
Sierra de Santa Marta and elsewhere have to
rely more than ever on local innovation,
adaptation, and diffusion as sources of
technical change. Nevertheless, even
autonomous action is hampered by the political
and organizational weakness of civil society in
the Sierra de Santa Marta. Farming
communities in the region are plagued with
internal problems of political conflict and
inadequate management skills. There are few
mechanisms through which farming










communities can offer alternatives of their own
or demand from the state policies and
programs which meet their needs. Thus, the
central challenge for implementation of the
alternative strategies for agricultural
intensification described in this paper lies in
the development of means to catalyze and
accelerate local processes of innovation and to
strengthen local capacity for self-management
and for informed negotiation with outside
agencies.

Cooperation between government agencies and
non-governmental organizations may have a
role to play in helping to create the conditions
for farmer-based agricultural development by
providing communities with access to
appropriate information, skills, and financial
resources. In 1991 and 1992, the PSSM and
CIMMYT undertook a campaign to facilitate
farmer access to cover crops and to strengthen
local capacity to generate and diffuse
appropriate agricultural technologies (Buckles,
Pare, and Arteaga 1995). This effort was
expanded in 1994 through cooperation with the
State Government of Veracruz to facilitate
farmer access to the inputs and information
needed to adopt contoured hedgerows as well.
While the initial PSSM-CIMMYT approach
emphasized farmer access to appropriate
information and seed, the coordinated strategy
with the state government also involved the
use of incentives to cover the estimated costs of
establishing the technologies and providing
technical assistance through farmer
paratechnicians. The following description
refers to the latter approach.

Project implementation began with a process of
farmer training in appropriate technologies
through local workshops. Contact farmers
known to regional institutions (municipal and
ejido authorities, NGOs, government agencies,


farmer organizations) were invited to a two-
day workshop in various locations during
which problems confronting regional farmers
were discussed using participatory exercises, a
few promising technical options were
presented (cover crops and contoured
hedgerows), and a program of incentives for
farmers was outlined. During the workshops
the farmers were trained in the use of the
technologies (including the A-frame for tracing
contour lines) by a small team of professional
extensionists.

The contact farmers were invited to take the
lead in establishing the technologies by
building their own A-frames and tracing the
contours for their own fields, work done
without compensation. They were also asked to
form a small group in their community which
in turn formally requested participation in the
program. The formation of self-selected small
groups provided a cohesive target for delivery
of the program and facilitated collective action
by participating farmers. The risk of diversion
of the program to more powerful members of
communities was reduced through the original
selection of contact farmers by a variety of
formal and informal institutions.

Contact farmers who had voluntarily and
successfully formed farmer groups and drawn
contours on their own fields were invited to
help implement the program; activities
included tracing contours in the fields of other
group members. These farmers, supported by
periodic visits and additional training by the
professional team, were accredited by the
program as paratechnicians (promotores) in their
communities and compensated for their work.
Promotores provided technical assistance about
establishing and maintaining contoured
hedgerows, using cover crops, and the contour
planting of maize. They also interviewed
farmers regarding their perspectives on the










technologies. Promotores participated in
evaluation workshops at the end of the season,
during which progress and problems
associated with the campaign were discussed
and plans made for future activities.

The 1994 campaign cost an average of US$ 172
per participating farmer. This estimate includes
the cost of the incentives, the technical
assistance, and the campaign coordination.
Incentives were the major cost, totaling some
US$ 102 per farmer. As compensation for their
work, farmers received 5 kg of mucuna seed
worth US$ 7; 5 kg of gliricidia seed worth US$
35; 50 kg of fertilizer (calcium superphosphate)
valued at US$ 10; a food package worth US$ 10
(beans, rice, cooking oil); and US$ 40 in cash.
With the exception of the mucuna seed, all of
these incentives were to facilitate establishment
of the contoured hedgerows, the most costly of
the technologies promoted.

The technical assistance was provided by the
paratechnicians at an estimated cost of US$ 40
per participating farmer. The paratechnicians
were paid US$ 30/ha to trace the contour lines
on each farmer's field; they were paid a total of
US$ 200 over the season to provide other forms
of technical assistance to the farmer group as a
whole. The latter payment averaged
approximately US$ 10 for each of the 1,457
farmers assisted.

The institutions coordinating the program
provided the staff time of three professional
extensionists and two assistants; periodic
assistance was also provided by several
administrators and technical people during a
six-month period. Implementation costs
included frequent visits to farming
communities by the professional team,
communications, office and administrative
support, the delivery of inputs such as the


fertilizer, and food disbursement. While
detailed records of these costs were not kept,
Table 2 estimates that the cost of campaign
coordination in 1994 averaged some US$ 30 per
participating farmer.

During the 1994 campaign, some 1,457 farmers
were organized into 56 groups, each led by a
paratechnician. More than 1,064 hectares of
contoured hedgerows were established by
farmers and paratechnicians and more than 9
tons of cover crop seed were planted on some
450 hectares of farmland. Farmer adoption of
the technologies resulting from this and earlier
campaigns has not been formally evaluated,
although farmer interest in the technologies
and expression of commitment to their
continued use has generally been strong
(Buckles, Pare, and Arteaga 1995). Work by
Soule (1995) suggests that approximately 55%
of the farmers receiving cover crop seed
through the earlier campaigns continued to
experiment with the technology the following
year.

The campaign did not consider incentives and
technical assistance needed to implement other
features of the alternative strategy proposed in
this paper, such as the efficient use of
commercial fertilizers, the conservation of crop

Table 2. Estimated costs of campaign coordination,
1994

Activity US$

Workshops (4) 2,800
Salaries, professional extensionists (3) 16,200
Salaries, field assistants (2) 7,200
Field coordination (transportation
and per diems for five coordinators) 8,870
Communications 1,670
Transport of seed, fertilizer, and food 3,500
Administration (10%) 4,024

Total 44,264










residues, increased maize planting densities,
the use of improved plant varieties, and
effective seed storage. Furthermore, without a
formal assessment of adoption rates resulting
from the campaign, it is impossible to estimate
the intensity and period of promotion needed
to achieve widespread adoption.

While generally successful, the 1994 campaign
encountered a number of problems and fell
well short of broader goals of organizational
development. Excessive and contradictory
bureaucratic procedures and late payment of
some incentives created tensions and
disagreements among farmers and between
farmers and the coordinators in various
communities. In addition, insufficient attention
was paid to key questions of farmer training in
communication skills, the impact of incentives
on farmer motivation, and the means to
enhance farmer participation in project
management (Buckles, Pare, and Arteaga 1995).

With these important qualifications in mind,
the campaign described above helps identify
the major elements of a potential
implementation strategy and the order of
magnitude of implementation costs. To be
effective, however, an implementation strategy
should not be limited to technical
considerations alone. In our view, the
appropriate intensification of agriculture also
requires concerted attention to broader issues
beyond the scope of this paper (e.g., regional
development plans, land use policies, and
grassroots community participation).


Conclusions

Farmers in the Sierra de Santa Marta have
traditionally grown maize in relatively
extensive slash-and-burn production systems


to meet their subsistence needs. While these
production practices were relatively sound and
sustainable in the past, they are no longer so in
the current environment. Shorter fallow
periods brought on by land use intensification
and insufficient technical adjustments by local
farmers to changing circumstances have
severely undermined the current maize
cropping system. Economic analyses presented
in the paper indicate in quantitative terms the
low profitability of the current maize
production systems. Quite simply, land, labor,
and capital are very poorly remunerated in
economic terms. Soil erosion, weed invasion
and the mining of soil fertility can be expected
to further reduce the viability of the cropping
system and to give rise to broader social costs
associated with out-migration and further
deforestation.

Despite problems of low profitability and land
degradation, maize production remains of vital
importance to smallholders concerned with
meeting their subsistence needs, and it is one of
the few productive uses of land and labor
currently available to them. Incomes in the
region are extremely low and employment both
on- and off-farm are very limited-conditions
that reinforce customary reliance on the
production of maize for household sustenance.
Growing maize is a survival strategy few
farmers can do without.

The use of plant nutrients available or
generated on the farm by cover crops in
combination with vegetative barriers and
limited external inputs could allow for the
appropriate intensification of maize-based
cropping systems in both the lowland and
upland zones of the Sierra de Santa Marta.
Farm-level analysis of the costs and benefits of
adoption indicate that while profitability
would remain low, the alternative cropping










system would be considerably more attractive
than the current one. Soil conservation
objectives both at the farm and watershed
levels might also be more effectively attained if
the practices perform as suggested by
qualitative experience from other similar
regions, thereby improving the sustainability of
the cropping system. In addition, some land
could be freed from maize production as a
result of a reduction in the maize area needed
to meet subsistence needs and a reduction in
the fallow area needed to maintain the
system-savings that amount to 1.33 ha per
household in the lowland zone and up to 4 ha
per household in the upland zone. This land is
potentially available for the production of more
commercially oriented crops or conversion to
secondary forest, thereby reducing the pressure
of the cropping system on the region's
remaining forest resources.


Economic analysis reveals that the timing of
adoption benefits is a significant constraint.
The proposed technologies require a
substantial initial investment by the farmer: a
discounted sum of US$ 140 per farm household
over five years by farmers in the upland zone
and a discounted sum of US$ 110 per farm
household over five years in the lowland zone.
After the initial investment period, the
proposed technologies may be self-supporting,
but cash-poor farmers would need help in
alleviating the investment cost. Both the
continuing economic crisis of the Mexican state
and the historical pattern of neglecting the
region's agricultural concerns suggest that
these needs will not be met through current
schemes. This paper provides an outline of a
farmer-based approach to implementing the
alternative strategies and a quantitative
assessment of the level of investment needed to
stimulate adoption and support a broader
process of sustained technology development
and adaptation in the region.











References


Andrle, R. 1964. A biogeographic investigation of
the Sierra de Tuxtla in Veracruz, Mexico.
Unpublished Ph.D. dissertation. Baton Rouge,
Louisiana: Louisiana State University.
Arias, R. (ed.) 1991. Primer diagn6stico para el
program de desarrollo integral de las Sierras en el
Estado de Veracruz: Sierra de Soteapan. Xalapa,
Veracruz: Fundaci6n Veracruz.
Baez-Jorge, F. 1973. Los Zoque-Popolucas. Mexico,
D.F.: Instituto Nacional Indigenista.
Barreto, H. 1994. Evaluation and utilization of
different mulches and cover crops for maize
production in Central America. In H.
Thurston, M. Smith, G. Abawi, and S. Kearl
(eds.), Slash/mulch: How Farmers Use It and
What Researchers Know About It. Ithaca, New
York: CIIFAD-CATIE. Pp. 157-167.
Bello, M. 1994. Evaluaci6n del program Fondo
Maicero de los Fondos Regionales de
Solidaridad. In Memoria del taller sobre las
political para una agriculture sustentable en la
Sierra de los Tuxtlas y Santa Marta, Veracruz, 3 y
4 de Marzo de 1994, Veracruz, Veracruz. Mexico,
D.F.: CIMMYT.
Blanco, J.L. 1995. Las diferentes caras del maiz entire
los popolucas de Soteapan. Unpublished
Manuscript. Mexico, D.F., and Xalapa,
Veracruz: CIMMYT and the Proyecto Sierra de
Santa Marta.
Blanco, J.L., D. Buckles, and H. Perales. 1994.
Avances del proyecto experimentaci6n con
variedades de maiz y selecci6n de semillas
locales con participaci6n campesina en la
Sierra de Santa Marta. Unpublished
Manuscript. Mexico, D.F., and Xalapa,
Veracruz: CIMMYT and the Proyecto Sierra de
Santa Marta.
Blom, F., and O. La Farge. 1926. Tribes and Temples.
New Orleans, Louisiana: Middle American
Research Series 1.
Boserup, E. 1965. The Condition of Agricultural
Growth. London, U.K.: Allen and Unwing.
Buckles, D. 1995. Farmer extension agent reports on
maize practices in the Sierra de Santa Marta,
Veracruz. Unpublished Manuscript. Mexico,
D.F.: CIMMYT.
Buckles, D., and H. Barreto. 1995. Intensificaci6n de
sistemas de agriculture tropical mediante
leguminosas de cobertura: Un marco conceptual.
CIMMYT/CIAT Internal Document. Mexico,
D.F.: CIMMYT.


Buckles, D., L. Pare, and L. Arteaga. 1995.
Experimentos hacia un modelo campesino
autogestivo de asistencia tecnica. Unpublished
paper. Mexico, D.F., and Xalapa, Veracruz:
CIMMYT and the Proyecto Sierra de Santa
Marta.
Buckles, D., and H. Perales. 1995. Farmer-based
experimentation with velvetbean: Innovation
within tradition. CIMMYT Internal Document.
Mexico, D.F.: CIMMYT.
Byerlee, D., and E. Hesse de Polanco. 1986. Farmers'
stepwise adoption of technological packages:
Evidence from the Mexican Altiplano.
American Journal of Agricultural Economics 68(3):
519-527.
Cervigni, R. 1995. The incremental cost of
biodiversity conservation in Sierra Santa
Marta, Southern Veracruz, Mexico: A PRINCE
case study. Internal draft document. PRINCE.
Washington, D.C.: GEF.
Chevalier, J., and D. Buckles. 1995. A Land Without
Gods: Process Theory, Maldevelopment and the
Mexican Nahuas. Halifax, Nova Scotia, and
London, U.K.: Fernwood Books and Zed
Books.
CIMMYT. 1988. La formulaci6n de
recomendaciones a partir de datos
agron6micos. Program de Economia. Mexico,
D.F.: CIMMYT.
Foster, G. 1942. A Primitive Mexican Economy. New
York, N.Y.: American Ethnological Society,
No. 5.
Garcia de Le6n, A. 1978. Pajapan: Un dialecto
mexicano del golfo. Mexico, D.F.: Instituto
Nacional de Antropologia e Historia,
Colecci6n Cientifica, No. 43.
Giller, K., and K. Wilson. 1991. Nitrogen Fixation in
Tropical Cropping Systems. Oxon, UK: CAB.
Gittinger, J.P. 1982. Economic Analysis of Agricultural
Projects (2nd edition). Baltimore, Maryland:
Johns Hopkins University Press.
Global Environment Facility/Proyecto Sierra de
Santa Marta, A.C./ Centro Internacional de
Mejoramiento de Maiz y Trigo. Forthcoming.
Desarrollo sustenable y conservacion de la
biodiversidad: Un studio de caso en la Sierra de
Santa Marta, Veracruz, Mexico. Raffaello
Cervigni and Fernando Ramirez, editors.
Mexico, D.F.: UNAM.
Guti6rrez Martinez, R. 1991. El medio fisico en la
Sierra de Santa Marta: Caracterizaci6n
regional. Unpublished Manuscript. Xalapa,
Veracruz: Proyecto Sierra de Santa Marta.











Gutierrez Martinez, R. 1995. La problematic de las
laderas en la Sierra de Santa Marta. In
Alternatives al manejo de laderas en Veracruz.
Xalapa, Veracruz: Secretaria del Medio
Ambiente, Recursos Naturales y Pesca/
Fundaci6n Friedrich Ebert.
Lobo Burle, M., A. Suhet, J. Pereira, D. Bouldin, D.
Lathwell, and W. Bowen. 1992. Legume Green
Manures: Dry-season Survival and the Effect on
Succeeding Maize Crops. Raleigh, N.C.: North
Carolina State University and the Soil
Management CRSP.
L6pez, J. 1993. Conservaci6n y productividad de
suelos en ladera de la Fraylesca, Chiapas.
Unpublished M.S. thesis. Monteccillo, Edo. de
Mexico: Colegio de Postgraduados.
Moscoso, B., and W. Raun. 1991. Evaluaci6n del
efecto de diferentes cultivos en relevo con
mafz. Paper presented at the XXXVIII Reuni6n
Anual del PCCMCA, Panama.
Oropeza, J.L., J. Rios, and N. Nicolas. 1994.
Componentes hidrol6gicos para evaluar la
terraza de muro vivo en el control de la erosi6n
hidrica en suelos de ladera. In Memoria del taller
sobre las political para una agriculture sustentable
en la Sierra de los Tuxtlas y Santa Marta, Veracruz,
3 y 4 de Marzo de 1994, Veracruz, Veracruz.
Mexico, D.F.: CIMMYT.
Par6, L., J. Chevalier, E. Velasquez, and D. Buckles
(eds.). Forthcoming. Hacia una estrategia de
desarrollo sustentable en la Sierra de Santa
Marta, Veracruz. Xalapa, Veracruz: Proyecto
Sierra de Santa Marta.
Pare, L., J. Aguero, and J.L. Blanco. 1994. Diagn6stico
de la producci6n del mafz en la Sierra de Santa
Marta. In Memoria del taller sobre las political para
una agriculture sustentable en la Sierra de los
Tuxtlas y Santa Marta, Veracruz, 3 y 4 de Marzo de
1994, Veracruz, Veracruz. Mexico, D.F.: CIMMYT.
Perales, H. 1992. Caracterizaci6n y valorizaci6n del
autoconsumo en la agriculture de los
popolucas de Soteapan, Veracruz. Unpublished
M.S. thesis. Monteccillo, Edo. de M6xico:
Colegio de Postgraduados.
Ponce, J. 1994. Politicas actuales de extension
agricola. In Memoria del taller sobre las political
para una agriculture sustentable en la Sierra de los
Tuxtlas y Santa Marta, Veracruz, 3 y 4 de Marzo
de 1994, Veracruz, Veracruz. Mexico, D.F.:
CIMMYT.
PRINCE, 1994. Proposed Case Study on the
Incremental Cost of Biodiversity Conservation
in the area of Sierra de Santa Marta, Mexico.
Internal draft document. Washington, D.C.:
GEF-PRINCE.


Ramirez, F. 1984. Plan conceptual para el manejo y
desarrollo de una Reserva de la Bi6sfera en la
Sierra de Santa Marta, Veracruz. Unpublished
Manuscript. Xalapa, Veracruz: Instituto
Nacional de Investigaci6n sobre Recursos
Bi6ticos.
Ramirez, F. 1991. Vegetaci6n y uso del suelo en la
zona Pajapan. Unpublished Manuscript. Xalapa,
Veracruz: Proyecto Sierra de Santa Marta.
Ramirez, F. 1992. Mapas de vegetacion y de deforestacion
de la Sierra de Santa Marta. Xalapa, Veracruz:
Proyecto Sierra de Santa Marta.
Rice, E., L. Godinez, and O. Erenstein. Forthcoming
(English and Spanish). The Farming Systems of
the Texizapan Watershed, Sierra de Santa Marta,
Veracruz. Mexico, D.F.: CIMMYT and PSSM.
Sanchez, P., and T. Cochrane. 1980. Soil constraints in
relation to major farming systems in tropical
America. In Priorities for Alleviating Soil-related
Constraints to Food Production in the Tropics. Los
Bafios, Philippines: IRRI.
Soule, M. 1995. Factors affecting farmer
experimentation with and assessment of
velvetbean: An economic analysis. Unpublished
paper. Mexico, D.F.: CIMMYT.
Stuart, J. 1978. Subsistence Ecology of the Isthmus
Nahuat Indians of Southern Veracruz, Mexico.
Unpublished Ph.D. dissertation. Riverside,
California: University of California.
Tasistro, A. 1994. La labranza de conservaci6n para la
Sierra de los Tuxtlas y Santa Marta. In Memoria
del taller sobre las political para una agriculture
sustentable en la Sierra de los Tuxtlas y Santa
Marta, Veracruz, 3 y 4 de Marzo de 1994, Veracruz,
Veracruz. Mexico, D.F.: CIMMYT.
Thurston, D. 1992. Sustainable Practices for Plant
Disease Management in Traditional Farming
Systems. Boulder, Colorado: Westview Press/
Oxford and IBH Publishing.
Wade, M., and P. Sanchez. 1983. Mulching and green
manure applications for continuous crop
production in the Amazon Basin. A;:..I ,'ii
Journal 75: 39-45.
Weischet, W., and C. Caviedes. 1993. The Persisting
Ecological Constraints of Tropical Agriculture.
London, U.K.: Longman Scientific and
Technical.
Z6fiiga, J.L., R. Camacho, S. Uribe, N. Nicolas, and A.
Turrent. 1993. Terrazas de muro vivo para
sustentar la productividad en terrenos agricolas
de ladera. In G. Diaz, J. Cruz, and C. Avila
(eds.), Memoria sexta reunion cientffica del sector
agropecuario yforestal del Estado de Veracruz,
Diciembre 9-10, Veracruz, Veracruz.










Appendix A


Crop Budgets


Appendix A presents the crop budgets
underlying the analysis in more detail.
However, before looking at the specific crop
budgets for each zone, a few observations are in
order in relation to the prices and yield
adjustment factors used.

Prices

Prices in the crop budgets are in US$. A major
reason for using US$ denominated prices is the
still uncertain impact the recent economic crisis
may have on actual and future domestic prices
through an unstable exchange and inflation
rate. It is assumed that US$ denominated prices
of most inputs and outputs will remain
relatively constant over time and thereby
present a more viable alternative to use as
constant 1995 prices over the time period
considered. A notable exception is the assumed
opportunity cost of labor (assumed equal to the
local wage rate of day laborers). Local labor
costs in Mexican peso terms are still relatively
constant (compared to one year ago and to
other costs), and thereby have declined
substantially in US$ terms. The sensitivity
analysis therefore specifically includes changes
in the assumed labor cost. Prices are indicative
only and should be treated with caution.

The on-farm price of maize is relatively high in
view of the subsistence orientation of the farm
household. The price is therefore not a
reflection of the actual on-farm sales price, but
more an estimate of the on-farm purchase price.
This is appropriate as most households under
the current maize production systems do not
meet subsistence requirements adequately (and
are thereby net-purchasers), whereas even in


the proposed maize production system,
production remains largely subsistence
oriented, with only an occasional sale of excess
production.

Yield Adjustment Factors

The data presented in the crop budgets assume
a rested field as the starting situation. However,
in the transition years, this is not the case for all
the subplots. The gradual transition from the
observed to the proposed case results in some
subplots starting in a previously used field (if
we assume the surplus plots to be put to
another use). Because these fields have a lower
productive potential, the yield response to the
package is adjusted downwards in the first two
transition years through a yield adjustment
factor. The adjustment factor for the first year is
based on the rate of yield decline in the
observed situation. The adjustment factor for
the second year is assumed to be half that of
the first year. In later years, the proposed
situation is presumed to prevail. In Figure 4
and Figure 5 the plots affected by yield
adjustment factors are labeled with an "a"
(adjusted). Table A-1 presents the yield
adjustment factors used in the analysis.

Table A-1. Yield adjustment factors

Yield adjustment
Zone Plot Year factor

Upland A3 4 90%
5 95%
A4 5 82%
6 91%
Lowland Al 3 71%
4 86%
A2/B1 4/5 60%
5/6 80%











Because labor use for harvest and shelling is
assumed to be proportional to yield, labor use
for these activities is also affected by the yield
adjustments.

Crop Budgets: Upland Zone

Table A-2 presents a detailed crop budget for
the actual cropping system (the without case)
for the model farm in the upland zone. The
budget is on a ha basis and presents data
annually for a specific field. It is assumed that a
field is cultivated for three years and is
subsequently fallowed for six years. The overall


implications of the budget are discussed in the
main text. Some details follow:

Field preparation: Assumes manual clearing of
acaual in the first year and therefore labor
demand is higher relative to subsequent
years.
Crop establishment: Sowing is manual with
espeque (dibble stick).
Weed control: Weeds become more problematic
in the subsequent years after clearing, and
consequently labor demand rises. Although
a number of farmers have started using
herbicides in the upland zone, such use is


Table A-2. Crop budget for the without case in the upland zone (ha basis)
Price Quantity Cost
Activity Unit (US$/units) (units/ha) (US$/ha)
Year 1 2 3 4 5-10 1 2 3 4 5-10


Field preparation Labor 2.5 20 16 16 50 40 40
0 0 0
Crop establishment Labor 2.5 6 6 6 15 15 15
Seed 0.21 16 16 16 3.4 3.4 3.4
Weed control Labor 2.5 25 30 35 62.5 75 87.5
Herbicide 5.5 0 0 0 0 0 0
Fertilizer applic. Labor 2.5 "0 0 0 0 0
Urea 11 "0 0 0 0 0
DAP 14 0 0 0 0 0
Doubling Labor 2.5 6 6 6 15 15 15
Harvest Labor 2.5 10.4 8.3 6.7 26.0 20.8 16.8
Shelling Labor 2.5 12.5 10 8 31.3 25.0 20.0
Additional mgt Labor 2.5 "0 0 0 0 0
Other "0 0 0 0 0
Depreciation implements Annuity 0 0 5 5 5
Subtotal labor 2.5 80 76 78 200 191 194








Net benefit (Return to land, US$/ha) -8 -59 -91
Cost price per kg (US$/kg) 0.17 0.20 0.25
Labor 0.16 0.19 0.24
Non-Labor no" 0.01 0.01 0.01
Return to labor and land (US$/ha) 192 132 104
Labor productivity (Kg maize output/day) 15.6 13.1 10.3
Return to labor day (and land) (US$/ld) 2.4 1.7 1.3










less widespread than in the lowland zone
for a variety of reasons. The budget for the
model farm in the upland zone assumes
manual weeding only.
Fertilizer application: External input use,
including chemical fertilizers, is limited in
the upland zone. The budget for the model
farm in the upland zone assumes no use of
chemical fertilizers.
Harvest: Harvesting labor is directly related to
yield, with an assumed manual harvesting
rate of four bags (30 kg of grain each) per
person per day.
Shelling: Shelling labor is directly related to
yield, with an assumed manual shelling
rate of 100 kg per day.
Maize yields: Yields decline over time as fields
become exhausted.
By-product: An estimated value for firewood
collected from the acaual at time of clearing.
Nevertheless, the opportunity value of
wood is low in view of large supply and
only limited demand for subsistence needs.
The actual estimate assumes that acaual
clearing generates a 10-day savings in time
normally spent collecting firewood.

Table A-3 presents a similar crop budget for the
improved cropping system (the with case) for
the model farm in the upland zone. Again the
budget is on a ha basis and presents data
annually for a specific field. It is assumed that a
field is cultivated for three years, and
subsequently fallowed for one year. However,
for a variety of reasons (see below), input and
output levels are different for subsequent
periods of use, and the budget therefore
presents data for both the first four years as
well as the subsequent four years (Years 6-9). In
even later years input and output levels similar
to the Years 6-9 are supposed to prevail. The
overall implications of the budget are discussed
in the main text. Some details follow:


Field preparation: Labor demands in the first year
are assumed to be similar to the without
case, except for the absence of burning. In
subsequent years, labor savings occur
because preparing the fields becomes
relatively easier.
Crop establishment: Labor needs for sowing are
assumed to be slightly higher than in the
without case due to the presence of crop
residue and mulch.
Weed control: Weeds become less problematic
over time as the cover crop, once fully
established, shades out most weeds thereby
substantially reducing labor needs for
weeding. Weed control remains manual
only.
Fertilizer application: Fertilizers are included in
the improved system. Whereas fertilizers
provide most of the N in the initial three
years (at an annual rate of 60 kg N per ha),
most of the subsequent N needs of the
maize crop are met by the cover crop (once
it is fully established). Consequently the N
application rate and the labor needs for its
application drop over time (stabilizing at
13.5 kg N per ha). P application rates are
assumed stable (35 kg P20, per ha).
Harvest: Harvesting labor is again directly
related to yield. In this case the assumed
manual harvesting rate is only 3.5 bags per
person per day due to the presence of the
cover crop, which slows harvesting.
Shelling: Shelling assumes the same rate as in
the actual system (100 kg per day).
Additional management of the cover crop: Labor
needs for sowing the cover crop in the first
year are assumed to be one-third those for
maize sowing (i.e., two days per ha).
Although the cover crop is sown with a
dibble stick at similar densities to maize, the
maize rows provide a guide to planting and
less care is needed to cover the cover crop
seed (Eilitta, personal communication).
Furthermore, an additional two days per ha











are required to prune mucuna where
needed (i.e., at times and places when it
competes with maize). In subsequent years,
labor for sowing the cover crop may
diminish as some natural reseeding will
occur. However, such savings will be offset
by an increased need to prune mucuna
plants that sprout in the field. In addition,
one day per season per ha is required for
cover crop seed collection; generally, seeds
can be gathered from plants left to mature


along the edge of fields. Other costs include
the initial seed requirements for the cover
crop (only for the first year as the farmer's
own seed is used in subsequent years).
Additional management of the hedgerow: Labor
needs for establishing a hedgerow in the
first year are assumed to total 22 labor days
(as estimated by Zufiiga et al. 1993); this
estimate includes seed collection, plus an
additional three days for reseeding
hedgerows). Additional installation costs


Table A-3. Crop budget for the with case in the upland zone (ha basis)
Price Quantity Cost
Activity Unit (US$/unit units/ha) (US$/ha)
Year 1 2 3 4 5 6 7-8 9 1 2 3 4 5 6 7-8 9

Field preparation Labor 2.5 18 14 14 16 14 45 35 35 40 35

Crop establishmeia.bor 2.5 7 7 7 7 7 17.5 17.5 17.5 "17.5 17.5
Seed 0.21 16 16 16 16 16 3.4 3.4 3.4 3.4 3.4
Weed control Labor 2.5 25 20 15 20 15 62.5 50 37.5 50 37.5
Herbicide 5.5 0 0 0 0 0 0 0 0 0 0
Fertilizerapplic. Labor 2.5 8 8 8 4 4 20 20 20 10 10
Urea 11 2 2 2 0 0 22 22 22 0 0
DAP 14 1.5 1.5 1.5 1.5 1.5 21 21 21 21 21
Doubling Labor 2.5 6 6 6 6 6 15 15 15 15 15
Harvest Labor 2.5 16.7 19 20.8 20.8 20.8 41.8 47.5 52 52 52
Shelling Labor 2.5 17.5 20 21.9 21.9 21.9 43.8 50.054.8 54.8 54.8
Additional mgt
cover crop Labor 2.5 5 5 5 5 5 12.5 12.5 12.5 12.5 12.5
Other 1 7.5 0 0 0 7.5 0 0 0 0
Additional mgt
hedgerow Labor 2.5 22 5 5 5 5 55 12.512.5 12.5 12.5
Other 1 30 0 0 0 30 0 30 0 0 0
Depreciation
implements Annuity 0 0 0 0 5 5 5 5 5
Subtotal labor 2.5 125 104 103 106 99 313 260 257 264 247
Subtotal non-labor 89 51 51 29 29
Total input 402 311 308 294 276

Maize grain consumption k@.14 1750 2000 2188 2188 2188 245 380 306 306 306
Other byproduct lump sum 1 0 0 0 0 0
Total 245 380 306 306 306
Net benefit (Return to land, US$/I a) -157 -31 -2 13 30
Cost price per kg (US$/kg) 0.23 0.16 0.14 0.13 0.13
Labor 0.18 0.130.12 0.12 0.11
Non-labor 0.05 0.03 0.02 0.01 0.01
Return to labor and land (US$/ha 156 229 255 277 277
Labor productivity (Kg maize outp t/day) 14.0 19.2 21.3 20.7 22.2
Return to Labor day (and land) ( S$/ld) 1.2 2.2 2.5 2.6 2.8










include one bag of DAP fertilizer and US$
16 for technical assistance in tracing
contours. Labor needs in subsequent years
mainly relate to hedgerow maintenance.
Maize yields: In the first year, yields are
assumed to increase a net 40% relative to
the without case, mainly as a direct response
to fertilizer application. In the second and
third years, yields are expected to rise a
further 20 and 15% relative to the first year
yields in the without case. These yield
increases are mainly a result of the other
measures (cover crops, residue
conservation, hedgerows) starting to take
effect, thereby increasing fertility and
available soil moisture. The presence of the
hedgerows reduces available crop area and
thereby potentially reduces yields.
However, this "productive loss" has already
been considered in the net yield increases.
Once the system is fully established, yields
are assumed to remain constant over time.

Crop Budgets: Lowland Zone

Table A-4 presents a detailed crop budget for
the actual cropping system (the without case) for
the model farm in the lowland zone. The
budget is on a ha basis and presents data for
each season (temporal and tapachole) annually
for a specific field. It is assumed that a field is
cultivated only for three seasons (1.5 years),
and is subsequently fallowed for five seasons
(2.5 years). The overall implications of the
budget are discussed in the main text. Some
details follow:

Field preparation: Assumes manual clearing of
grassy acaual for the temporal season in the
first year and therefore labor demand is
higher than for the second year. Manual


clearing for the tapachole season only
involves reweeding/clearing of the temporal
maize field since tapachole maize is sown as
an intercrop between the doubled temporal
maize.
Crop establishment: Labor needs for sowing the
tapachole intercrop are assumed to be two-
thirds of the temporal labor needs because
the doubled maize plants used during the
temporal season can serve as guides.
Weed control: Weeds become more problematic
in the subsequent years after clearing;
consequently, labor demand rises in the
second temporal season. Labor demand for
weeding in the tapachole season is relatively
low as a result of limited rainfall and the
presence of doubled temporal maize stalks.
Labor demand for weeding in the lowland
zone is lower than in the upland zone due
to the combined use of herbicides and
manual weeding (the upland zone budget
assumes manual weeding only).
Fertilizer application: Fertilizer application is
relatively common in the lowland zone, but
is mainly limited to the temporal season due
to production risks in the tapachole season.1
Annual application rates at the model farm
are assumed to average 55 kg N per ha and
23 kg P20, per ha.
Doubling: In contrast to temporal maize, tapachole
maize is not doubled.
Harvest: Harvesting labor is directly related to
yield, as in the without case in the upland
zone (four bags per person per day).
Shelling: Shelling assumes the same rate as in
the without case in the upland zone (100 kg
per day).
Maize yields: Yields decline over the temporal
seasons as fields become exhausted. Yields in
the tapachole season average only 500 kg/ha.


1 Fertilizer use in the lowland zone was relatively common in the years prior to the 1995 temporal season. The economic crisis
during 1995 limited fertilizer use during that cycle, but it is expected that fertilizer use will rebound in the without case in
the subsequent years.
















In
IC




















-- dd-d
--











I C L M r CT- -- T
S0.. O
E.














CEL



CE
c0
'- 0. -. -h ... o








C i








o- 0 c



04 -O
E 00















E w ocLmrN 0
04
^CIS
0.0









E 0.







U) U) as t'D _0


D I M co coi co
o (D
(0 CM I-

















r- = 0) 0 0
0 1Z oj 0I0f)








c 0. as C: 0-
0- Z CM (Z-_ --- -
Z) aD I-

L. 2 01 0 CZ 6 W3Wo
o^ U o >, r o o








UT M QI co< 00a20 1 C C 0 -








44










By-product: In contrast to the upland zone,
acauales in the lowland zone are generally
grassy and thus produce only a negligible
amount of firewood.

Table A-5 presents a similar crop budget for the
improved cropping system (the with case) for
the model farm in the lowland zone. Again, the
budget is on a ha basis and presents data for
each season (temporal and tapachole) annually
for a specific field. It is assumed that a field is
cultivated annually, with seasonal fallows (two
tapachole fallows, one temporal fallow) spread
out over the four-year cycle. However for a
variety of reasons (see below) input and output
levels are different for subsequent periods of
use, and the budget therefore presents data for
both the first four years as well as the
subsequent four years (Years 6-9). In even later
years, input and output levels similar to the
Years 6-9 are supposed to prevail. The overall
implications of the budget are discussed in the
main text. Some details follow:

Field preparation: Labor demands for the first
year are assumed to be similar to those in
the without case, except for the absence of
burning. In subsequent years labor savings
occur as a result of relatively easier
preparation (slashing) of fields with cover
crops (both for temporal and tapachole
seasons).
Crop establishment: Labor needs for sowing are
assumed to be slightly higher than in
without case due to the presence of crop
residue and mulch.
Weed control: Weeds become less problematic
over time as the cover crop, once fully
established, shades out most weeds, thereby
substantially reducing labor and herbicide
needs for weeding in both seasons.


Fertilizer application: Fertilizers levels in temporal
season are increased in the improved system.
Whereas fertilizers provide most of the N in
the initial two years, most of the subsequent
N needs of the maize crop are supplied by
the cover crop (once it is fully established).
Consequently, the application rate and the
labor demand for application drop over time.
Application rates are assumed to total 110 kg
N per ha and 46 kg P0O, per ha in the early
years, and only 60 kg N per ha and 35 kg
PO, per ha in the later years.
Harvest: Harvesting labor is directly related to
yield, as in the with case for the upland zone
(3.5 bags per person per day).
Shelling: Shelling assumes the same rate as in the
without case in the upland zone (100 kg per
day).
Additional management of the cover crop:
Establishment and maintenance costs of the
cover crop are assumed to be similar to those
in the upland zone. Labor required for
resowing cover crop for temporal fallow is
assumed to total only one day per ha.
Additional management of the hedgerow:
Establishment and maintenance costs of the
hedgerow are assumed to be similar to costs
in the upland zone.
Maize yields: In the first year, yields are assumed
to increase a net 40% relative to the without
case, mainly as a direct response to increased
fertilizer application. In the second and third
years, yields are expected to rise a further 30
and 10% relative to the first year in the
without case. These yield increases are mainly
a result of the other measures starting to take
effect (cover crops, residue conservation,
hedgerows), thereby increasing fertility and
available soil moisture. Once the system is
fully established, yields are assumed to
remain constant over time.




















o L1C\ 10
C\j N T-


LO NC\ Lo


LO LO
- C\J
CM CM


CO LO


o LO LO LO CMLoCCPP
C) ( C C C -OCO
O D Co


C) r It- CM (O L
C) LO LO


o OLOLQo Loj W
r- CD D- M


co
O0


CO O CO
CM


CO NO C


- CM -


LO LO
CM 1


C\J


Lq :o) t OD


o LO C LO
10 COOC)

C\j CD O C CO\J
C\j C()


CLO\j ,o CO
cm C


LO o o 0 C
C\j It
tme mem
cu*


LO CO


NN
Co


C Lo (0 t
C\LO
CM


,t C\J (0 C co
CM CO)
C\j C\j
Schir
cuc


Co IOLO N CO
C)- Mo M t w oCO .
C\J OD m


LO LO







0
LN


LOLO C\ 0
SCMCm


CO CO
C\J C\J


OD OD


CO) CO)
LO LO
CO CO


CO) CO)
CO) CO)
CO) CO)
m a


j- C-


C\j CO CLO C N 10
66 C) C -o
a o
000~'C


- 66

O CO

0 0
CY Y)C
C m 1 7
C) C


C\j

C\j 0 C
a'C\
C\Jom
o o



(SaC


o LO CO CM
c? 666 C\J CM
0 c; CM
a rea a
ddd Cac


LO C\JI mC C -
17 0C C -


-'_ 4- 4


C\J C\j C C\j C\JC C\ J C \J C\


o o o o oo o0 o C r


f.Q i -- _l n r -aC C

.o E E -o
.. 0 0
I a Ni C O ^ig



m3 2 QT t Q:0> 0
E 0 S O 8 E (


oE

0-
E




"
C. 0 -

E
mr
.N cn 0

oOF-


0)
B
(I)


a


0-
C-)
.o
-j
CL)
o o
0 -Q


0
~r


r -
0 E 0


C c
co 0






C 0
M _J
-i cr



















LO LO
- C\J
C\j C\j


o \oCLo
C0 1~~ S *


LO 1 C\J LO


LO


C? C?
0C
O4 O


)LO O Ct) JOJCJ- LO CD
CO m "C C c C mCO
CO co co


O O C) O .N CC ) O Io
C m "C C C OCO
CO CO CO


SL. O IC LO LO C O1D99
Co m I-- N- OCO
CO cCO CO


CO o CO


CO: No CO
^ CV


CO


- 0
CO CO


~- NOC\JLqNC\JO CO~
CU C\


C CLO
CM


C CLO
CM


C\j C\J







o MO COC C\j
CO CY>



C L0CO CO(
C\j COC C\J
SC\j CO







"0 0COLO CO








LOa 0 C LO CO
C\j (COC C\J
CM CO
ommj (


10 (0







Ct
Ct


10 10
0


LO. LO -


10O 10 10;
0


LO LO LO LO LO L L L L O LO
C\j C) O"C L8 C .C\ C. C\J,- C\J,-


O O0 ncoma0000 Ca) Ca)
-0 -0" Q 05 0 ctn 0 -0 0 0 -CB 0 -C


- _
co c a)m mm m m m

C C/) M 0 0 :
a t
0 E 6

a) -r 0 0')~a
co o a g> C
co
0r 0 co 7a 7a 2:
v 0

a) 70 N S 0 0


II 3: | 0 1 -C _0 -0 Q
) LL QI/)< E < -


0

o -



E E

a n n


CO CO


CO) CO)
LO LO
CO CO


LO LO
03 0


CO a
LO LO
03 0


2






En


086


0 CO C\J C\J I- L
000C CC- -
coors~


0 o CO C
00 -C\


0 "00 C" CM

C\J





S CO CO s
0C0 0 CM C\J


C .


000 CJ "c\j


0) _

C: CO



C >, co

S0-0
0 no0 -n




o a~ca)
o amw


cn
Cis3

8%Z


ci



C

003







T

Z

o* c
0 D
in
< ,



o
B
(u


I-Q >5


1 I
CT>CT












Appendix B


Labor Use


Labor is an important farm resource that
potentially limits intensification efforts.
Therefore, determining the labor requirements
in both the with and without cases helps to
assess the potential for intensification.
Furthermore, it is useful to distinguish between
labor requirements by operation and by month
(Gittinger 1982, p. 102). The crop budgets in
Appendix A highlighted labor use by
operation. Appendix B presents the temporal
distribution of labor use in more detail.


Table B-1 is the key used to assign labor
requirements per activity for respective
months. April has been chosen as the start of


the cropping year because field preparation
activities for the main temporal season generally
start in this month. Table B-2 presents the labor
distribution by year and month on a ha basis
for the upland zone. Because the with case
exhibits higher annual totals, it is not too
surprising it also generally presents slight
increases in the monthly requirements.
However, due to differences in acreages, a
farm-level comparison offers a more realistic
means of determining eventual labor
bottlenecks. Table B-3 presents the labor
distribution by year and month on a farm basis
for the upland zone. At the farm level, annual
labor requirements are similar for both the with


Table B-1. Assumed temporal distribution of activities

Month
Zone Crop Activitity Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar


Upland & lowlandMaize temporal Field preparation
Crop establishment
Weed control & fertilizer
Doubling
Harvest (see footnote)
Shelling

Upland & lowlandCover intercrop Sowing
Pruning
Seed collection


Lowland


Cover sole cropSowing
Seed collection


Upland & lowlandHedgerow


Lowland


Establishment (1st yeafy*************
Reseeding (1st year)
Maintenance (2nd year do'f***


Maize tapacholeField preparation
Crop establishment
Weed control & fertilizer
Harvest
Shelling


Harvest in lowland zone generally concentrated in the early months; in the buffer zone harvest generally extended over the longer


r********


************** *
************


*****


******


******
***********











and without situation. The with situation gives a The following observations in relation to the


more uniform distribution of labor
requirements over the year. Especially in the
busy months of May-July the with situation
alleviates the original labor peak.

Tables B-4 and B-5 present similar data for the
lowland zone, where annual labor
requirements are higher for the with case. As a
result, monthly labor requirements per ha and
for the farm as a whole are generally higher
too. Nevertheless, the with case only represents
a slight increase in labor requirements in the
first month of the busy period of May-July, and
actually reduces labor requirements in the
subsequent two months.


tables are in order:

* Monthly labor requirements are indicative only.
Requirements have thus been rounded to the
nearest day As annual totals were calculated
before rounding, these may differ slightly from
the sum of monthly (rounded) totals.
* Annual totals in the lowland zone occasionally
differ from the respective annual totals
reported in the crop budgets. This is mainly the
result of the tapachole season, which causes an
overflow of labor (for shelling) into the
subsequent cropping year (see Table B-l).
* Column maxima before rounding are
highlighted in bold.


Table B-2. Labor distribution by year and month on ha basis, upland zone

Labor days/Month
Year Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Total
Without case 2 7 13 12 13 6 0 6 4 4 5 5 5 80
3 5 11 13 16 7 0 6 3 3 5 4 3 76
4 5 11 15 17 9 0 6 3 3 3 3 2 78
5-10 Fallow
With case 2 11 24 20 17 10 1 7 7 8 7 7 6 126
3 5 14 14 14 9 1 7 8 9 8 8 7 104
4 5 14 13 11 8 1 7 9 10 9 8 8 103
5 Fallow
6 5 16 13 12 8 1 7 9 10 9 8 8 106
7-8 5 14 12 9 7 1 7 9 10 9 8 8 99
9 Fallow


Table B-3. Labor distribution by year and month on farm basis, upland zone

Labor days/Month
Year Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Total
Without case 2 14 26 24 26 12 0 12 8 8 11 10 9 160
3 10 22 26 32 14 0 12 6 6 10 8 7 153
4 10 22 30 34 18 0 12 6 6 7 7 4 155
With case 2 13 25 22 22 11 1 10 8 8 9 8 8 143
3 11 25 24 24 13 1 10 9 10 10 10 8 153
4 11 26 24 21 14 2 11 12 13 12 11 10 164
5 11 26 24 21 14 2 11 11 12 11 11 10 162
6 8 22 20 19 13 2 11 13 14 13 12 11 154
7 8 22 19 16 12 2 11 14 15 13 12 12 154
8-> 8 22 19 15 11 2 11 14 15 13 12 12 152
Note: For the without case, years after year 5 are repetitions of years 2-4.











Table B-4. Labor distribution by year and month on ha basis, lowland zone

Labor days/Month
Year Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Total

Without case 2 6 12 11 12 5 0 6 14 12 11 5 8 102
3 8 13 13 13 7 0 6 4 4 4 3 3 78
4-5 Fallow

With case 2 10 20 18 14 8 1 7 10 10 7 7 6 117
3 4 15 13 11 8 1 7 12 12 8 8 8 107
4 4 15 12 9 7 1 7 18 20 13 8 15 129
5 4 8 1 0 0 0 1 8 9 4 0 9 44
6 9 19 12 9 7 1 7 12 13 9 8 8 114
7 4 15 12 9 7 1 7 12 13 9 8 8 105
8 4 15 12 9 7 1 7 18 20 13 8 16 130
9 5 9 1 0 0 0 1 8 9 4 0 9 45


Table. B-5. Labor distribution by year and month on farm basis, lowland zone

Labor days/Month
Year Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Total

Without case 2-> 9 17 16 17 8 0 8 12 10 10 5 7 120

With case 2 7 15 13 13 6 0 6 13 11 10 6 7 107
3 9 20 18 17 9 1 9 12 12 9 7 8 131
4 9 21 19 16 10 1 9 13 13 9 7 9 136
5 7 19 15 11 8 1 7 14 15 9 6 11 124
6 7 19 13 10 7 1 7 16 17 11 8 13 128
7 7 19 12 9 7 1 7 17 18 12 8 13 131
8-> 7 19 12 9 7 1 7 17 18 12 8 14 131












Appendix C


Cost Benefit Analysis Results


Table C-1. Farm-level implications over time of intensifying model farm in upland zone (farm basis)

Discount rate 15%

Year-> 1 2 3 4 5 6 7 8 9 10 1-5 6-10 TOTAL

Total labor Id 155 160 153 155 160 153 155 160 153 155 783 776 1559
Labor costs US$ 338 302 251 222 199 165 146 131 108 96 1311 646 1958
Non-labor costs US$ 15 13 11 10 8 7 6 5 5 4 56 28 84
Total input costs US$ 352 315 262 232 207 172 152 136 113 100 1368 674 2041
Total output kg 1600 2500 2000 1600 2500 2000 1600 2500 2000 1600 10200 9700 19900
Gross benefit US$ 195 302 184 128 199 121 84 131 80 55 1008 471 1479
Net benefit US$ -158 -12 -78 -104 -8 -51 -68 -5 -34 -45 -359 -203 -562

Total labor Id 155 143 153 164 162 154 154 152 152 152 777 763 1539
Labor costs US$ 338 269 251 236 204 168 144 124 108 94 1299 638 1936
Non-labor costs US$ 15 40 49 55 48 29 21 14 13 11 206 87 293
Total input costs US$ 352 309 300 291 252 197 165 138 120 105 1504 725 2229
Total output kg 1600 2125 2375 2881 2761 3098 3281 3281 3281 3281
Gross benefit US$ 195 244 219 231 192 187 173 150 131 114 1080 754 1835
Net benefit US$ -158 -65 -81 -60 -61 -9 8 12 10 9 -424 30 -395

Total labor Id 0 -17 0 9 2 2 -2 -8 -1 -4 -6 -13 -19
Labor costs US$ 0 -33 0 14 6 3 -2 -7 -1 -2 -13 -8 -21
Non-labor costs US$ 0 27 38 45 39 21 14 9 8 7 150 59 209
Total input costs US$ 0 -5 38 59 45 24 13 2 7 4 137 51 188
Total output kg 0 -375 375 1281 261 1098 1681 781 1281 1681 1543 6523 8065
Gross benefit US$ 0 -59 35 103 -7 66 88 19 251 58 72 283 355
Net benefit US$ 0 -53 -4 43 -52 42 76 17 44 54 -65 233 168
B/C ratio NR -1079% 91% 173% -15% 272% 699% 887% 726% 1330% 52% 559% 189%













Table C-2. Sensitivity analysis of farm-level implications of intensifying model farm in upland zone (farm basis)


= Quant- -- ,itpr ha re-discound ls.). SU.BTTA TOT I


Year-> 1 2


3 4 5 6 7 8 9 10 1-4 5-10 1-10


Discount rate: 10%
NET (WITH- WITHOUT)
Total labor Id 0 -17 0 9 2 2 -2 -8 -1 -4 -6 -13 -19
Labor costs US$ 0 -36 0 17 7 4 -2 -10 -1 -4 -12 -13 -24
Non-labor costs US$ 0 30 43 54 49 28 20 13 12 11 176 82 259
Total input costs US$ 0 -6 44 71 56 32 17 3 10 7 165 70 234
Total output kg 0 -375 375 1281 261 1098 1681 781 1281 1681 1543 6523 8065
Gross benefit US$ 0 -64 39 123 -8 87 121 28 76 91 90 402 492
Net benefit US$ 0 -58 -4 52 -64 55 103 25 66 84 -75 332 257
B/C ratio NR -1079% 91% 173% -15% 272% 699% 887% 726% 1330% 54% 577% 210%
Discount rate: 20%
NET (WITH- WITHOUT)
Total labor Id 0 -17 0 9 2 2 -2 -8 -1 -4 -6 -13 -19
Labor costs US$ 0 -30 0 12 5 2 -1 -5 -1 -2 -13 -6 -19
Non-labor costs US$ 0 25 33 38 32 17 11 6 5 4 128 43 172
Total input costs US$ 0 -5 34 50 36 19 9 2 5 3 115 38 152
Total output kg 0 -375 375 1281 261 1098 1681 781 1281 1681 1543 6523 8065
Gross benefit US$ 0 -54 30 87 -5 51 66 14 35 38 58 204 261
Net benefit US$ 0 -49 -3 36 -42 33 56 12 30 35 -57 166 109
B/C ratio NR -1079% 91% 173% -15% 272% 699% 887% 726% 1330% 50% 543% 171%


Opportunity cost labor: US$ 1.5 / day
NET (WITH -WITHOUT)
Total labor Id 0 -17 0 9 2 2 -2 -8 -1 -4 -6 -13 -19
Total labor Id 0 -17 0 9 2 2 -2 -8 -1 -4 -6 -13 -19
Labor costs US$ 0 -20 0 8 3 2 -1 -4 0 -1 -9 -5 -14
Non-labor costs US$ 0 27 38 45 39 21 14 9 8 7 150 59 209
Total input costs US$ 0 8 38 53 42 23 13 5 7 5 141 54 195
Total output kg 0 -375 375 1281 261 1098 1681 781 1281 1681 1543 6523 8065
Gross benefit US$ 0 -51 35 103 3 66 88 26 51 58 89 290 379
Net benefit US$ 0 -59 -3 49 -39 44 75 21 44 53 -52 236 185
B/C ratio NR 667% 91% 192% 8% 290% 662% 531% 697% 1092% 63% 539% 195%
Opportunity cost labor: US$ 3.5 / day
NET (WITH- WITHOUT)
Total labor Id 0 -17 0 9 2 2 -2 -8 -1 -4 -6 -13 -19
Labor costs US$ 0 -46 0 20 9 5 -2 -9 -1 -3 -17 -12 -28
Non-labor costs US$ 0 27 38 45 39 21 14 9 8 7 150 59 209
Total input costs US$ 0 -19 38 65 48 26 12 -1 7 3 133 48 181
Total output kg 0 -375 375 1281 261 1098 1681 781 1281 1681 1543 6523 8065
Gross benefit US$ 0 -66 35 103 -17 66 88 13 51 58 54 277 331
Net benefit US$ 0 -48 -4 37 -65 40 77 13 44 55 -79 229 151
B/C ratio NR -357% 90% 157% -35% 255% 739%-2532% 759% 1700% 41% 582% 183%


Initial yield increase: 30%
NET (WITH -WITHOUT)
Total labor Id 0 -19 -3 4 -3 -4 -9 -15 -8 -11 -21 -47 -68
Labor costs US$ 0 -35 -5 7 -1 -4 -8 -12 -6 -7 -34 -37 -70
Non-labor costs US$ 0 27 38 45 39 21 14 9 8 7 150 59 209
Total input costs US$ 0 -8 33 52 39 18 6 -3 2 0 116 23 138
Total output kg 0 -438 219 1014 4 778 1330 430 930 1330 799 4797 5596
Gross benefit US$ 0 -65 20 81 -25 47 70 3 37 46 11 203 215
Net benefit US$ 0 -58 -13 29 -63 29 64 7 35 46 -104 181 77
B/C ratio NR -847% 61% 157% -64% 267% 1132% -96% 1748%41833% 10% 901% 155%
Initial yield increase: 50%
NET (WITH- WITHOUT)
Total labor Id 0 -16 3 14 7 8 5 -1 6 3 8 21 29
Labor costs US$ 0 -30 5 22 12 10 5 -1 4 2 9 20 29
Non-labor costs US$ 0 27 38 45 39 21 14 9 8 7 150 59 209
Total input costs US$ 0 -3 43 67 51 31 19 8 12 9 159 79 237
Total output kg 0 -313 531 1548 519 1417 2033 1133 1633 2033 2286 8248 10534
Gross benefit US$ 0 -52 49 124 11 86 107 35 65 70 132 364 496
Net benefit US$ 0 -49 6 57 -40 54 88 28 53 62 -26 285 258
B/Cratio NR -1695% 113% 185% 22% 274% 559% 454% 545% 814% 83% 461% 209%












Table C-3. Farm level implications overtime of intensifying model farm in lowland zone (farm basis)

Discount rate 15%
US$ are discounted values
Year -> 1 2 3 4 5 6 7 8 9 10 1-5 6-10 TOTAL

Total labor Id 120 120 120 120 120 120 120 120 120 120 120 120 120
Labor costs US$ 261 227 197 172 149 130 113 98 85 74 1007 500 1507
Non-labor costs US$ 74 64 56 49 42 37 32 28 24 21 286 142 428
Total input costs US$ 335 292 253 220 192 167 145 126 110 95 1292 643 1935
Total output kg 1933 1933 1933 1933 1933 1933 1933 1933 1933 1933 1933 1933 1933
Gross benefit US$ 235 205 178 155 135 117 102 88 77 67 907 451 1358
Net benefit US$ -100 -87 -76 -66 -57 -50 -43 -38 -33 -28 -385 -191 -577

Total labor Id 120 110 131 137 127 128 131 131 131 131 625 653 1278
Labor costs US$ 261 209 219 202 164 140 123 107 93 81 1054 545 1599
Non-labor costs US$ 74 73 80 69 53 36 27 23 20 18 349 124 473
Total input costs US$ 335 282 299 270 217 176 150 131 114 99 1403 669 2072
Total output kg 1933 1920 2226 2478 2400 2848 3053 3103 3103 3103 10958 15210 26167
Gross benefit US$ 235 203 205 198 167 172 161 142 123 107 1009 706 1715
Net benefit US$ -100 -79 -94 -72 -50 -3 11 11 10 8 -395 37 -357

Total labor Id 0 -10 11 17 7 8 10 11 11 11 25 52 77
Labor costs US$ 0 -18 21 30 14 10 10 9 8 7 48 44 92
Non-labor costs US$ 0 9 24 20 11 -1 -5 -4 -4 -3 63 -18 45
Total input costs US$ 0 -10 45 50 25 9 5 5 4 4 111 26 137
Total output kg 0 -13 293 545 467 915 1120 1170 1170 1170 1292 5544 6836
Gross benefit US$ 0 -1 27 44 32 55 59 54 47 40 102 255 357
Net benefit US$ 0 8 -18 -6 7 46 54 49 42 37 -9 229 219
B/C ratio % NR 15% 59% 87% 129% 613%1272% 1111% 1111%1111% 92% 969% 259%













Table C-4. Sensitivity analysis of farm level implications of intensifying model farm in lowland zone (farm basis)

~ ~ ~ ii 4 -4 -:


Year-> 1 2 3


4 5 6 7 8 9 10 1-5 6-10 1-10


SSNSIIVIY AALYIS DICON RAT


Discount rate: 10%
NET (WITH WITHOUT
Total labor
Labor costs
Non-labor costs
Total input costs
Total output
Gross benefit
Net benefit
B/C ratio


Id 0 -10 11 17 7 8 10 11 11 11 25 52 77
US$ 0 -20 25 36 18 14 13 13 12 11 58 63 122
US$ 0 10 27 24 14 -2 -7 -6 -6 -5 74 -26 48
US$ 0 -11 52 60 32 12 6 7 6 6 133 37 170
kg 0 -13 293 545 467 915 1120 1170 1170 1170 1292 5544 6836
US$ 0 -2 31 52 41 72 80 76 69 63 122 362 484
US$ 0 9 -21 -8 9 60 74 70 63 57 -11 325 314
% NR 15% 59% 87% 129% 613% 1272% 1111% 1111% 1111% 92% 980% 285%


Discount rate: 20%
NET (WITH WITHOUT)
Total labor Id 0 -10 11 17 7 8 10 11 11 11 25 52 77
Labor costs US$ 0 -17 19 26 12 8 7 7 6 5 39 32 71
Non-labor costs US$ 0 8 21 17 9 -1 -4 -3 -3 -2 55 -13 42
Total input costs US$ 0 -9 40 42 20 7 3 3 3 2 94 19 113
Total output kg 0 -13 293 545 467 915 1120 1170 1170 1170 1292 5544 6836
Gross benefit US$ 0 -1 24 37 26 43 44 38 32 26 86 183 268
Net benefit US$ 0 8 -16 -5 6 36 40 35 29 24 -8 164 156
B/C ratio % NR 15% 59% 87% 129% 613% 1272% 1111% 1111% 1111% 91% 958% 238%
SENSITIVITY ANLSSOPOTNT CS 0AO


Opportunity cost labor:
NET (WITH WITHOUT
Total labor
Labor cost
Non-labor costs
Total input costs
Total output
Gross benefit
Net benefit
B/C ratio


US$ 1.5 / day
TI


Id 0 -10 11 17 7 8 10 11 11 11 25 52 77
US$ 0 -11 12 16 7 6 6 6 5 4 24 26 50
US$ 0 9 24 20 11 -1 -5 -4 -4 -3 63 -18 45
US$ 0 -2 36 36 18 4 1 1 1 1 88 8 96
kg 0 -13 293 545 467 915 1120 1170 1170 1170 1292 5544 6836
US$ 0 -1 27 44 32 55 59 54 47 40 102 255 357
US$ 0 1 -9 7 15 51 58 52 46 40 14 247 261
% NR 61% 75% 120% 182% 1272% 8110% 4841% 4841% 4841% 116% 3193% 372%


Opportunity cost labor: US$ 3.5 / day
NET (WITH WITHOUT)
Total labor Id 0 -10 11 17 7 8 10 11 11 11 25 52 77
Labor costs US$ 0 -26 31 44 22 15 14 13 11 10 71 63 134
Non-labor costs US$ 0 9 24 20 11 -1 -5 -4 -4 -3 63 -18 45
Total input costs US$ 0 -17 55 64 33 14 9 9 7 6 134 45 179
Total output kg 0 -13 293 545 467 915 1120 1170 1170 1170 1292 5544 6836
Gross benefit US$ 0 -1 27 44 32 55 59 54 47 40 102 255 357
Net benefit US$ 0 16 -28 -20 -0 42 50 45 39 34 -33 210 178
B/C ratio % NR 8% 49% 68% 100% 404% 690% 628% 628% 628% 76% 571% 199%


Initial yield increase: 30%
NET (WITH WITHOUT)
Total labor Id 0 -11 9 13 2 2 4 5 5 5 13 20 32
Labor costs US$ 0 -20 18 24 8 4 4 4 3 3 29 17 46
Non-labor costs US$ 0 9 24 20 11 -1 -5 -4 -4 -3 63 -18 45
Total input costs US$ 0 -11 42 44 19 2 -1 -1 -1 -1 93 -1 92
Total output kg 0 -60 178 337 230 613 790 827 827 827 686 3885 4570
Gross benefit US$ 0 -6 16 27 16 37 42 38 33 29 53 178 231
Net benefit US$ 0 5 -25 -17 -3 35 43 39 34 29 -40 179 139
B/C ratio % NR 55% 40% 62% 85% 1548% -2816% -5531%-5531% -5531% 57% -2E+04 252%


Initial yield increase: 50%
NET (WITH WITHOUT)
Total labor
Labor costs
Non-labor costs
Total input costs
Total output
Gross benefit
Net benefit
B/C ratio %


Id 0 -9 13 21 11 14 17 18 18 18 36 85 121
US$ 0 -17 25 37 21 17 16 15 13 11 66 71 137
US$ 0 9 24 20 11 -1 -5 -4 -4 -3 63 -18 45
US$ 0 -8 49 56 31 16 11 10 9 8 129 53 183
kg 0 33 408 753 703 1217 1450 1512 1512 1512 1898 7203 9102
US$ 0 4 38 60 49 74 76 69 60 52 150 332 482
US$ 0 12 -12 4 17 58 66 59 51 45 21 278 299
NR -44% 76% 107% 156% 472% 715% 672% 672% 672% 116% 622% 264%


I=


ITn











New Papers from the Natural Resources Group


Paper Series
95-01 Land and Livelihoods: Patterns of Rural Development in Atlantic Honduras
D. Buckles and G. Sain

96-01 Meeting South Asia's Future Food Requirements from Rice-Wheat Cropping Systems: Priority
Issues Facing Researchers in the Post-Green Revolution Era
P. Hobbs and M. Morris

96-02 Soil Fertility Management Research for the Maize Cropping Systems of Smallholders in Southern
Africa: A Review
J.D.T. Kumwenda, S.R. Waddington, S.S. Snapp, R.B. Jones, and M.J. Blackie

96-03 Genetic Diversity and Maize Seed Management in a Traditional Mexican Community: Implications
for In Situ Conservation of Maize
D. Louette and M. Smale

96-04 Indicators of Wheat Genetic Diversity and Germplasm Use in the People's Republic of China
N. Yang and M. Smale

96-05 Low Use of Fertilizers and Low Productivity in Sub-Saharan Africa
W. Mwangi

96-06 Intensificaci6n de sistemas de agriculture tropical mediante leguminosas de cobertura: Un marco
conceptual
D. Buckles and H. Barreto

96-07 Intensifying Maize-based Cropping Systems in the Sierra de Santa Marta, Veracruz
D. Buckles and 0. Erenstein

96-08 In Situ Conservation of Crops and Their Relatives: A Review of Current Status and Prospects for
Wheat and Maize
Gerard J. Dempsey


Reprint Series
96-01 Evaluating the Potential of Conservation Tillage in Maize-based Farming Systems in the Mexican
Tropics
O. Erenstein










ISSN: 1405-2830


SInternational Maize and Wheat Improvement Center
Centro Internacional de Mejoramiento de Maiz y Trigo
SLisboa 27, Apartado Postal 6-641, 06600 M6xico, D.F., M6xico




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs