• TABLE OF CONTENTS
HIDE
 Front Cover
 Title Page
 Copyright
 Table of Contents
 List of Figures
 Executive summary
 Introduction
 Sampling and data collection
 Characterization of farm house...
 Maize varietal knowledge
 Maize and seed relief program...
 Effectiveness of maize seed assistance...
 Factors determining the recycling...
 Lessons learned from maize seed...
 Concluding remarks
 Annex 1
 Annex 2
 Back Cover






Title: Assessment of the effectiveness of maize seed assistance to vulnerable farm households in Zimbabwe
CITATION PDF VIEWER PAGE IMAGE ZOOMABLE PAGE TEXT
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00077489/00001
 Material Information
Title: Assessment of the effectiveness of maize seed assistance to vulnerable farm households in Zimbabwe
Physical Description: Book
Language: English
Creator: Langyintuo, Augustine S.
Publisher: International Maize and Wheat Improvement Center (CIMMYT)
Publication Date: 2007
 Subjects
Subject: Africa   ( lcsh )
Farming   ( lcsh )
Spatial Coverage: Africa -- Zimbabwe
Africa
 Record Information
Bibliographic ID: UF00077489
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: African Studies Collections in the Department of Special Collections and Area Studies, George A. Smathers Libraries, University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: isbn - 970-648-148-6

Downloads

This item has the following downloads:

faoseedassistance ( PDF )


Table of Contents
    Front Cover
        Front cover
    Title Page
        Page i
    Copyright
        Page ii
    Table of Contents
        Page iii
    List of Figures
        Page iv
    Executive summary
        Page 1
        Page 2
        Page 3
    Introduction
        Page 4
    Sampling and data collection
        Page 5
    Characterization of farm households
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
    Maize varietal knowledge
        Page 13
        Page 14
        Page 15
    Maize and seed relief program activities
        Page 16
        Page 17
    Effectiveness of maize seed assistance to vulnerable households
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
    Factors determining the recycling of relief seed by households
        Page 24
        Page 25
    Lessons learned from maize seed relief to vulnerable groups in Zimbabwe
        Page 26
    Concluding remarks
        Page 27
    Annex 1
        Page 28
        Page 29
    Annex 2
        Page 30
    Back Cover
        Back cover
Full Text


Assessment of the Effectiveness of
Maize Seed Assistance to
Vulnerable Farm
Households in Zimbabwe

Augustine S. Langyintuo and Peter Setimela


a'
"YM
,*YTM-
-4













Assessment of the Effectiveness of


Maize Seed Assistance to Vulnerable Farm

Households in Zimbabwe







Augustine S. Langyintuo and Peter Setimela


International Maize and Wheat Improvement Center (CIMMYT)


April 2007






Acknowledgements

This publication was made possible with financial and technical support from the Food and Agricultural
Organization (FAO) Regional Office in Harare, Zimbabwe, through Michael Jenrich and his colleagues.
The ideas expressed here are those of the authors and do not necessarily reflect those of FAO. Many
thanks to the NGOs and local authority representatives for providing the lists of farmers in the selected
areas for interviews. We are grateful to the enumerators for administering the questionnaires and the
farmers for accepting to be interviewed. We thank John Dixon, Director of CIMMYT's Impacts Targeting
and Assessment Unit, for valuable comments on the text. The authors acknowledge the editorial support
of Mike Listman, CIMMYT, and the layout and production assistance of designers Miguel Mellado and
Eliot Sanchez.










The International Maize and Wheat Improvement Center, known by its Spanish acronym, CIMMYT
(www.cimmyt.org), is an international, not-for-profit research and training organization that, together with partners
in over 100 developing countries, works to increase food security, improve the productivity and profitability of
farming systems, and sustain natural resources in the developing world. The center's outputs and services include
improved maize and wheat varieties and cropping systems, the conservation of maize and wheat genetic resources,
and capacity building. CIMMYT belongs to and is funded by the Consultative Group on International Agricultural
Research (CGIAR) (www.cgiar.org), and also receives support from national governments, foundations,
development banks, and other public and private agencies.

International Maize and Wheat Improvement Center (CIMMYT) 2007. All rights reserved. The designations
employed in the presentation of materials in this publication do not imply the expression of any opinion whatsoever
on the part of CIMMYT or its contributory organizations concerning the legal status of any country, territory, city,
or area, or of its authorities, or concerning the delimitation of its frontiers or boundaries. CIMMYT encourages fair
use of this material. Proper citation is requested.

Correct citation: Langyintuo, A.S., and P. Setimela. 2007. Assessment of the effectiveness of maize seed assistance
to vulnerable farm households in Zimbabwe. Mexico, D.F.: CIMMYT.

Abstract: The publication describes outcomes of a study to assess the effectiveness of a large-scale crop seed relief
effort in Zimbabwe during 2003-07. Aims of the effort, which was supported by the British Department for
International Development (DfID) and coordinated by the Food and Agricultural Organization (FAO) regional office
in Harare, included broader diffusion of open-pollinated maize varieties (OPVs), as opposed to hybrids. Based on
the findings of the study, the authors recommend that, to increase benefits to vulnerable groups, participants in such
efforts should effectively disseminate information on selecting and recycling seed, supported by training and field
demonstrations, and should target relatively well-endowed farmers initially. Recommendations also included
promotion of OPVs by commercial seed companies and developing a simple, farmer-friendly system for naming
varieties.

ISBN: 970-648-148-6
AGROVOC descriptors: Agricultural development; Technical aid; Farm income; Food production; Maize; Open
pollination; Hybrids; Zimbabwe
AGRIS category codes: E10 Agricultural Economics and Policies
F30 Plant Genetics and Breeding
Dewey decimal classification: 338.16689


Printed in Mexico.









Table of contents

E x ecutiv e su m m ary ...................................................................... ................... ............. .. 1
Introduction .. ........................................................................................................ ..... 4
Sam pling and data collection ....................................................................... ...................... 5
C characterization of farm hou seholds .............................................................................................. 6
D em graphic characteristics of households..................................................................... ..... 6
Household access to productive assets ........................... ....... .................................... 7
W health indices ................................. ... ..................... ........... ........... ...... 9
Distribution of productive assets by wealth category ................................................... 12
M aize varietal know ledge .......................................................... ............ ............ .. 13
M aize seed relief program activities ............................................................................. ...... ...... 16
Maize seed distributed to beneficiaries...................... ...... ....................... 16
Effectiveness of maize seed assistance to vulnerable households.............................................. 18
Factors determining the recycling of relief seed by households............................................. 24
Lessons learned from maize seed relief to vulnerable groups in Zimbabwe............................ 26
The flow of information from NGOs to program beneficiaries on seed selection ................ 26
The relative importance of factors influencing farmers' seed selection decisions ................ 26
Concluding remarks ........................ .......... .................... 27
A nnex 1 ................................................ 28
A nnex 2 .................................. ...... ..... .............. 30


List of Tables

Table 1: Descriptive statistics of survey districts in Zimbabwe .............................................. 6
Table 2: Distribution of family size by province and gender .................. .................... 7
T able 3 : A access to productive assets ................ ........................................................................... 8
Table 4: Distribution of farm size by district and gender ...................................... .................... 8
Table 5: Selected households wealth indicators by wealth category in Zimbabwe.................. 12
Table 6: Sources of maize seed planted by farmers in Zimbabwe ............................ ............. 14
Table 7: Comparison of OPV and hybrids by farmers in terms of selected characteristics ......... 15
Table 8: Information flow between NGO and farmers.............................. ..................... 16
Table 9: Maize varieties distributed under the PRP program over the last three years (%)......... 17
Table 10:Average quantity of seed in (kg) received by beneficiary farmers............................ 17
Table 11: OPV maize seed beneficiaries' attitudes toward seed selection........................... .. 21









List of Figures


Figure 1: Distribution of cropped land to crops at provincial level in Zimbabwe........................ 9
Figure 2: Distribution of households according to wealth groups ............................................ 10
Figure 3: Wealth index ranking of households in selected districts in Zimbabwe ...................... 11
Figure 4: Probability distribution of households by wealth categories.................. .......... 11
Figure 5: Proportional distribution of assets by wealth group..................... .............. 12
Figure 6: Proportional distribution of productive assets by wealth group......... .............. 13
Figure 7: Distribution of relief seed type by wealth group........... .......................... 18
Figure 8: Access to and use of information by seed relief beneficiaries .................................. 19
Figure 9: Information on seed type and beneficiaries' preferences for the seed ...................... 19
Figure 10: Flow and use of OPV seed selection information by beneficiaries........................... 22
Figure 11: Access to recycling information by OPV maize seed beneficiaries.......................... 22
Figure 12: Seed selection behavior of beneficiaries based on information and knowledge......... 23
Figure 13: Information on and seed selection behavior of beneficiaries by wealth group........... 23
Figure 14: Information on and seed selection behavior of beneficiaries by gender ..................... 24

List of Maps

Map 1: Map of Zimbabwe showing survey districts............. ... ................. ........................5









Executive summary


Introduction

The British Department for International Development (DflD) implemented a protracted relief
program (PRP) to help vulnerable households in Zimbabwe improve their food security and
livelihoods. One of five components of the PRP was the distribution of seed of major food crops
(maize, sorghum, cowpeas, pearl millet and groundnuts) to vulnerable farm households in
selected districts through non-governmental organizations (NGOs) and coordinated by the Food
and Agricultural Organization (FAO) regional office in Harare. Over 25,000 households in the
Makoni, Hurungwe, Shurugwe, Chivi, Seke, and Gwanda districts in the Manicaland,
Mashonaland West, Midlands, Masvingo, Mashonaland East, and Matebeleland South provinces,
respectively, benefited from the relief seed between 2003/04 and 2006/07 crop seasons.

The proportion of maize OPVs versus hybrids distributed increased from 54% when the program
started to 95% in the third year for two main reasons. Firstly, OPVs seed are relatively cheap and
out-perform hybrids when cultivated under marginal conditions without complementary inputs
such as fertilizer. Secondly, because the yield reduction effect of recycled OPVs is far lower than
for hybrids, farmers could recycle the seeds. The objective of this study was to assess the
effectiveness of the seed distribution component of the PRP. Specifically the study investigated
whether or not NGOs advised beneficiaries to select, store and re-use maize OPV seed; assessed
the retention of relief OPV maize seed within target farming communities; based on the findings,
reported on lessons learned; and recommended ways to improve seed assistance programs. Using
structured questionnaires, data were collected from 100 households in each of the six districts
listed above, 70% beneficiaries and the remaining 30% non-beneficiaries. Three questionnaires
were rejected during analysis on the basis of incomplete information.

Profile of a typical farm household

There are no significant differences in the typologies of households between beneficiaries and
non-beneficiaries of the PRP. The average household size is 6.5 members with comparative
distribution within age groups across the six districts. Of the 597 households interviewed, 213
are headed by females, 86% of whom are widowed compared to only 6% of their male
counterparts. Cultivated farm size averages 1.7 ha, over 60% of which is planted to maize OPVs
and hybrids. About half of the farmers recycle OPV and hybrid maize. Most farmers think OPV
seeds are relatively cheaper than hybrids but agreed that the former are hardly available on the
local markets. Two-thirds think that OPVs are inferior to hybrids in terms of yield, resistance to
field and storage pests as well as tolerance to low soil fertility or drought.

Unlike wheelbarrows, productive assets such as pairs of bullock (or donkeys), bicycles and
harrows are less popular among farmers. The proportions of farmers owning ox-ploughs is about
twice that owning bullocks and donkeys put together because some farmers reportedly disposed
of their draft animals for cash to meet immediate household needs or could not maintain them
when economic conditions became very difficult.









Maize seed distributed to beneficiaries


Of the 417 beneficiaries interviewed, 26% received maize seed in 2003/04, 52% in 2004/05 and
76% 2005/06. The average quantity of seeds distributed varied from 5 to 16 kg per beneficiary
depending on year and location. Over the three years, between 50 and 70% of the beneficiaries
thought the seeds were always delivered on time for planting. The dominant OPVs distributed
were ZM421, ZM423, ZM523 and ZM521, while among hybrids, SC 530, SC513, SC501,
SC413, SC405, SC403, and SC401 dominated. More than 50% of the beneficiaries in the first
year indicated that given cash and seed availability1 on the market, they will be willing to
purchase similar varieties suggesting that farmers like the varieties distributed. The proportion,
however, declined over time with the shift from hybrids to OPVs.

Effectiveness of maize seed assistance to vulnerable households

The effectiveness of the relief seed distribution was assessed by analyzing the flow of
information from NGOs to beneficiary households on maize seed types distributed, the need to
select, store and re-use OPV maize seed in subsequent years, and how much OPV maize seed
distributed is being retained in the selected communities. Results from the survey suggest that
less than half of the first-year beneficiaries were informed of the type of seeds to be provided.
The proportion (and hence number of beneficiaries) increased to a little more than 60% by the
third year. Information on what an OPV is was limited to the fact that it is a maize variety that
can be recycled. In general, the proportion of beneficiaries informed to select seeds increased
from 55% the first year to 62% in the third. Over the three-year period, 25% received such
information more than once, and 75% only once. In terms of carrying out field demonstrations to
teach farmers how to select or store seed, less than 50% of them benefited in the three
consecutive years.

Notwithstanding the fact that the varieties seemed to meet the preferences of most beneficiaries
in terms of performance in the field and food qualities, less than 50% of them actually selected
seed. Females appear more willing to select compared with males. Similarly, recycling is more
common among the relatively richer households than the very poor. Results from regression
analysis show that the probability that a farmer initially without any experience with recycling
OPV maize seed would recycle will increase by 24% with just a year's experience provided the
seed meets his/her preferences. The strategies used by different NGOs appear to have differential
impacts on beneficiaries' selection behavior.

Lessons learned from maize seed relief to vulnerable households

The survey results clearly show that, over the three-year period of the PRP, less than 10% of the
beneficiaries ever participated in deciding on the type of maize seeds to be distributed. The
program emphasized OPV seed distribution that would allow farmers to recycle but it is for the
same reason that seed companies are reluctant to produce and market such seeds. The results,
however, suggest that both OPVs and hybrids are recycled.
The proportion of beneficiaries informed to select, store, and re-use OPV seed by participating
NGOs increased from 55% the first year to 62% by the third, much greater that the proportion

1 The unavailability of similar varieties on the local markets was of concern to more than 80% of beneficiaries.









taught how to select. Aggregated over the years, the flow of seed selection information was
gender insensitive. Each year, fewer than half of beneficiaries informed to select actually did so,
although females appear more aggressive in selecting seed. If seed selection information is
complemented by teaching farmers how to select, the proportion of those selecting increases.
Factors observed to be important in driving seed selection include information on the need to
select, past experience with seed selection, and if the variety meets the preferences of the
beneficiary. In addition, some NGOs appear more effective than others in convincing their
beneficiaries to select seeds.

Concluding remarks

Based on the findings of the study, the following recommendations are made to improve seed
assistance to vulnerable groups in Zimbabwe to ensure greater spillover effects.

Experience with OPV maize seed recycling is important in influencing farmers' recycling
decisions. Ways to extend similar interventions for a few more years should be explored.
In this way, many more farmers would have experienced handling OPV maize seed and
hence be recycling seed, ensuring greater spillover effects within the communities.
Given that the relatively "well-endowed" farmers are more willing to recycle OPV seed,
targeting them can potentially ensure large-scale spillovers.
NGOs should endeavor to ensure that all beneficiaries be provided with seed selection
information. They should follow up the information with field-level
training/demonstration of seed selection practices during the crop growth period. This
could help foster greater adoption of selection than simply providing information.
Seed selection information should emphasize the need to select only OPV maize, since
farmers are inclined to select any type of variety.
Participating NGOs should share their seed selection information strategies so that less
successful ones can learn from those able to ensure that large proportions of their
beneficiaries select seed.
Beneficiaries should be given the opportunity to select the varieties they want, because
the probability of recycling is higher if a variety meets their preferences.
Seed companies should promote OPVs: farmers recycle both OPVs and hybrids but, if
given a choice, they will purchase fresh seed every year, thereby creating an OPV
market.
A simple varietal naming system should be developed to help farmers remember the
names of the varieties they receive and to distinguish between hybrids and OPVs.










Introduction


The recent economic downturn in Zimbabwe severely impoverished many households. As a
result, the majority of them could no longer afford essential commodities. Farmers found prices
of farm inputs (such as fertilizer and seed) beyond their means thereby making them vulnerable
to food insecurity and their livelihoods threatened. To help vulnerable households improve their
food security and livelihoods, the British Department for International Development (DflD)
implemented the protracted relief program (PRP) for 3 years (2003/04-2006/07). One of the five
components of the PRP was the distribution of seed of major food crops (maize, sorghum,
cowpeas, pearl millet and groundnuts) to vulnerable farm households in selected districts through
non-governmental organizations (NGOs) coordinated by the Food and Agricultural Organization
(FAO) regional office in Harare (Annex 1).

Seed of both hybrid and open pollinated varieties (OPVs) of maize, the dominant crop produced
and consumed in the country2, constituted a major part of the program. The proportion of OPVs
versus hybrids increased from 54% in 2003/04 when the PRP started to 87% in 2004/05 and 95%
in 2005/06 for two main reasons. Firstly, most of the vulnerable farm households grow maize in
marginal areas under sub-optimal conditions and lack money to purchase complementary inputs
recommended for optimal performance of hybrids. Moreover, hybrid seed costs over 20% more
than that of OPVs, implying a reduced assistance per household and thereby the opportunity to
reach more households without increasing funding. Secondly, for lack of cash to purchase seed,
most farmers "recycle" seed-that is, they save grain from the harvest and sow it as seed the
following year. In that regard, OPV maize offers the best choice: the yield reduction from
recycling freshly purchased seed the following season is only 5%, compared with 30% for
hybrids3. What the PRP hoped to achieve was that, at the end of the program, farmers would
sustainably source their seed, with spill-over effects in the communities. This was ultimately
going to depend on several factors, including committed farmers and NGOs who would inform
them how to select, store, and re-use seed of OPVs rather than hybrids, which were once the
predominant type of maize grown in Zimbabwe.

To date, over 25,000 beneficiaries in six districts have received OPV maize seed, but it was not
know if they had indeed begun to recycle it. The objective of this study was to assess the
effectiveness of the PRP in this regard, specifically:
* Investigate whether or not NGOs advised beneficiaries to select, store and re-use maize
OPV seed distributed to them.
* Assess the retention of relief OPV maize seed within target farming communities.
* Report on lessons learned and recommend ways to improve seed assistance programs.
While providing a deeper analysis of the overall impact of such assistance programs, the
study was expected to generate insights that would improve the effectiveness of the
distribution of relief seed among vulnerable farm households in Zimbabwe.

2 Byerlee, D. and C. Eicher, eds. 1997. Africa's Emerging Maize Revolution. Lynne Rienner, Boulder, Co. Food
Composition Tables, 1987. Technical Center for Agriculture and Rural Cooperation, Waginingen Agricultural
University, Netherlands.
3 Pixley, K., Banziger, M., 2002. Open pollinated maize varieties: A backward step or valuable option for farmers.
International Maize and Wheat Improvement Center (CIMMYT), Harare Zimbabwe.










Sampling and data collection

This study was a collaborative effort between the FAO regional office in Harare, Zimbabwe and
the International Maize and Wheat Improvement Center (CIMMYT) in Harare, Zimbabwe. FAO
provided financial and some technical support to CIMMYT to implement the study. The
selection of survey locations was influenced by the PRP related activities coordinated by the
former. In the past three years, the FAO coordinated the distribution of relief seed to selected
beneficiaries in the Manicaland, Mashonaland West, Midlands, Masvingo, Mashonaland East,
and Matebeleland South provinces of Zimbabwe. The field work therefore concentrated on the
Makoni, Hurungwe, Shurugwe, Chivi, Seke, and Gwanda districts in the above-named
provinces, respectively (Map 1). Farm households were sampled from all the wards that
participated in the relief seed distribution over the last three years. In each district, 70% of the
sampled households were beneficiaries and the remaining 30% non-beneficiaries. From the FAO
records (Annex 1), 16 NGOs distributed 420 tons of maize OPV seed to 25,232 farm households
in 11 wards in the Makoni (5,020), Hurungwe (6,642), Shurugwe (1,881), Chivi (3,538), Seke
(2,544) and Gwanda (5,607) districts during the 2004/05 and 2005/06 cropping seasons.


)


*1C


_ 'if
U ~


-I, -,
-, -1i
* -I~ .t.i


-,
-Ii ( ~-


ji


i ....... I: '


In each district, the councilor or local authority representative provided the list of households in
the selected areas, which was used as the sampling frame. Seed relief beneficiary households
were identified by participating NGOs. For a representative sample of beneficiary households,
the number of farm households selected from each ward and per participating NGO was
proportional to the number of beneficiaries in the ward assisted by the NGO (See Annex 1
column 6).


r


j









Characterization of farm households


Demographic characteristics of households

The descriptive statistics of selected farm households presented in Table 1 suggest that the
average household size is 6.5, with comparative distribution of members within age groups
across the six districts. Over 50% of the households have members ranging from 4 to 8 (Table 2).
Within that family size range, female headed households appear to have more members than
their male counterparts. The reverse is true for household sizes larger than 8 members.

Table 1: Descriptive statistics of survey districts in Zimbabwe
District
Makoni Hurungwe Shurugwe Chivi Seke Gwanda
Variable (n= 100) (n= 100) (n= 100) (n=99) (n=99) (n=99)
Household size (number)
Total members1 6.2(2.9) 6.5(2.6) 5.8(2.9) 6.9(3.1) 6.1 (2.4) 7.8(3.2)
Above 50 years 0.9(0.9) 1.0(0.8) 0.8(0.8) 0.9(0.9) 0.8(0.8) 0.9(0.8)
Males (16-50 years) 1.5 (1.1) 1.5 (1.1) 1.4 (1.2) 1.6 (1.3) 1.4 (1.1) 1.6 (1.4)
Female (16-50 years) 1.4(1.3) 1.6(1.2) 1.5(1.5) 1.6(1.4) 1.6(1.1) 1.9(1.3)
Males (9-15 years) 0.6(0.8) 0.6(0.8) 0.6(0.8) 0.8(1.0) 0.7(0.9) 1.0(1.1)
Females (9-15 years) 0.6 (0.9) 0.6 (0.8) 0.6 (0.7) 0.7 (0.9) 0.6 (0.8) 0.9 (1.2)
Children under 8 years 1.2(1.2) 1.2(1.2) 0.9(1.1) 1.2(1.2) 1.0(1.1) 1.6(1.5)
Age of HH head (years)2 50(15) 54(18) 55(17) 51 (16) 50(14) 53(13)
Female HH heads (%) 30 45 31 37 32 38
Marital status ofHH head (%)
Single 0 3 1 0 0 2
Married 74 51 57 61 66 62
Divorced 3 1 6 2 2 2
Widowed 23 45 36 37 32 34
Widowed by gender (%)
Females 70 91 87 92 91 84
Males 3 2 13 5 4 3
Educational level ofHH head (%)
Illiterate 3 24 8 8 0 3
Primary 56 49 42 56 54 47
Secondary 41 26 46 36 46 48
Post secondary 0 1 4 0 0 1
Association membership (%) 72 26 66 71 75 34
Note: 'In parentheses are the standard deviations; 2H = household









Table 2: Distribution of family size by province and gender
Gender of Family size range (number of members)
District household head <2.1 2.1 -4 4.1 -6 6.1 -8 8.1 10 >10
Female (n=30) 3 23 43 13 3 13
Makoni Male (n=70) 4 29 29 21 14 3
Female (n=45) 4 20 47 20 4 4
Hurungwe Male (n=55) 2 16 25 33 15 9
Female (n=31) 10 16 42 13 13 6
Shurugwe Male (n=69) 14 25 23 25 7 6
Female (n=37) 8 24 19 27 8 14
Chivi Male (n=62) 0 21 31 18 19 11
Female (n=32) 6 28 41 16 3 6
Seke Male (n=67) 1 24 34 19 16 4
Female (n=38) 0 8 45 24 8 16
Gwanda Male (n=61) 2 5 26 30 18 20

Each household is headed by the most elderly person aged about 52 years, 36% of whom are
females. Of the 213 female household heads interviewed, 86% of them are widowed compared to
6% of their male counterparts. Whereas in Seke all the household heads have some formal
education up to standard six, 24% of those in Hurungwe are illiterate (Table 1). The highest
educational level attained by about 4% of household heads in Shurugwe is post secondary. About
half of all the farmers interviewed belong to at least a farmers' association.

Household access to productive assets

Table 3 indicates that farm households are endowed with varying levels of different assets. Farm
sizes range from 1.4 ha in Seke to about 3 ha in Gwanda. The majority of households in all but
Seke district own between 1 and 3 hectares with few female headed households owning above 3
ha (Table 4). In Seke, the high population pressure on farm land limits the maximum farm size to
only 2 ha for more than 80% of them.

Maize is the dominant food crop in all districts (Annex 2). It occupies more than half of the
cultivated land in all districts except Chivi (Table 3 and Figure 1). Farmers also keep livestock,
predominantly fowls, as a risk management strategy (Table 3). In times of crop failure, which
sometimes leads to loss of seed stock, livestock are sold to purchase grain for home consumption
and seed for planting.

Productive assets such as pairs of bullock (or donkeys as draft animals), bicycles and harrows are
less popular among farmers than wheelbarrows, which appear common across the districts.
Whereas more than half of the farmers in Shurugwe own scotch carts, the opposite is true for
those in the other districts. The proportion of farmers owning ox-ploughs is about twice that
owning bullocks and donkeys put together, apparently because some farmers reportedly disposed
of their draft animals (but not the ploughs) for cash to meet immediate household needs or
because they could not maintain them when economic conditions became very difficult for them.









Table 3: Access to productive assets
Makoni Hurungwe Shurugwe Chivi Seke Gwanda
(n=100) (n=100) (n=100) (n=99) (n=99) (n=99)
Farm land ownership
Total farm (ha) 2.0(1.1) 3.5 (2.5) 2.8(2.6) 3.4(2.6) 1.4(0.8) 2.9(1.5)
Cultivated farm (ha) 1.4 (0.8) 2.1 (1.6) 1.5 (1.1) 2.4 (1.8) 0.9 (0.4) 1.9 (1.1)
Maize area (%) 60 76 64 44 66 69
Livestock numbers
Cattle 2.0(3.5) 1.1 (2.1) 3.3 (5.5) 2.6(3.3) 1.0(2.1) 2.4(3.4)
Small ruminants 1.9(2.6) 2.1(3.0) 1.3(1.7) 2.7(4.1) 0.7(1.0) 4.4(3.7)
Fowls 8.2 (7.5) 6.9 (6.8) 9.2 10.1) 8.5 (6.8) 5.5 (6.9) 8.1 (5.9)
Ownership of at least one ...
Pair bullock 39 27 43 43 18 42
Donkey 0 4 7 43 1 57
Scotch cart 34 27 55 36 20 48
Bicycle 34 20 21 21 14 30
Radio set 33 37 31 22 27 39
Wheel barrow 60 38 69 43 35 68
Ox-plough 62 69 61 69 37 85
Harrow 11 14 34 20 12 22
Note: In parentheses are standard deviations


Table 4: Distribution of farm size by district and gender
Gender of HH Farm size range (ha)
District head < 1.01 1.01 2.0 2.01 3.0 3.01 4.0 4.01 5.0 >5.0
Female (n=30) 20 40 27 3 7 3
Makoni Male (n=70) 33 37 23 4 1 1
Female (n=45) 9 36 29 13 9 4
Hurungwe Male (n=55) 2 27 29 15 5 22
Female (n=31) 6 55 26 3 3 6
Shurugwe Male (n=69) 20 32 26 10 3 9
Female (n=37) 14 30 24 14 5 14
Chivi Male (n=62) 8 27 31 6 10 18
Female (n=32) 66 28 3 3 0 0
Seke Male (n=67) 46 43 7 1 1 0
Female (n=38) 13 24 39 13 5 5
Gwanda Male (n=61) 7 26 39 16 5 7




































Figure 1: Distribution of cropped land to crops at provincial level in Zimbabwe

Wealth indices

Households are endowed with varying levels of different assets, as noted earlier, each of which
could potentially contribute to their wealth statuses. In Zimbabwe as in most developing
countries, smallholder farmers are usually cash-strapped and have limited access to credit for
varied reasons and therefore have to rely on their productive assets to chart a route out of
poverty. To be able to assess how the wealth levels of households influenced farmers' access to
technology and in particular information on relief seed selection and re-use, it was important to
generate composite wealth indices using the farmers' assets. Such an approach will also give an
indication of whether the distribution of seed actually targeted the very poor in the communities.
In other words, the process could indirectly serve to validate the targeting processes adopted by
the NGOs.

From the whole sample, an estimated 55% of the households are poorly endowed with a mean
index of -0.79. The well-endowed, on the other hand, have a mean index of 0.92. Figure 3 shows
that at the district level, the proportion of households within the well-endowed category in
Shurungwe appear larger than in all other districts because of the smaller probability of getting a
household below the sample mean of zero (42%) compared with all others, especially Seke (over
70%).


Beans Cowpeas
4% 4%
Bambarabeans
7% \

Groundnuts
16%


Rice aize
1% 61%
Sorghum
5%

Millet
2%









Using the mean index values for the poorly-endowed and well-endowed households as lower and
upper cut-off points, respectively, Figure 4 shows the probability distribution of households in
three categories: below the lower cut-off point of -0.72, between the lower and upper cut-off
points of 0.92 and above the upper cut-off point. Clearly, the Seke district has over 50% of its
households below the poorly-endowed mean compared with all other districts. This means that if
a development program aims to improve the livelihoods of the very vulnerable groups in the
society, Seke district should take priority over the others.


Household


Figure 2: Distribution of households according to wealth groups































-2.0 -1.0 0.0 1.0 2.0 3.0 4.0

-- Makoni Hurungwe --- Shurugwe -- Chivi Seke --- Gwanda

Figure 3: Wealth index ranking of households in selected districts in Zimbabwe


100%

90%
80%
0.42
70%
60%
0.46 0.64
50% 0.37 0.42 0.63
40%

30%
20%

10%
0%
Makoni Hurungwe Shiruigwe Chivi Seke Gwanda

Makoni Hurungwe Shurugwe Chivi Seke Gwanda

Prob (Between poorly- and well- endowed) 0.46 0.65 0.38 0.43 0.43 0.64

Figure 4: Probability distribution of households by wealth categories




11









Distribution of productive assets by wealth category


After estimating the wealth indices it was found expedient to examine the distribution of selected
assets between the poorly- and well-endowed households. There did not seem to be any
difference in the distribution of beneficiaries and non-beneficiaries of relief seed within wealth
groups. A striking difference in the distribution of gender of household head within the wealth
categories was, however, observed (Table 5). About 64% of the female-headed households fall
under the poorly-endowed households compared with only 48% of the male-headed households.
The proportion of farmers within the well-endowed group owning physical assets such as pair of
bullocks, ploughs, harrows, scotch carts, etc., far exceeds that within the poorly-endowed group
(Figure 5). In the case of land use, 63% of the cultivated land is owned by the 45% of well-
endowed households (Figure 6). The numbers of cattle, goats, sheep and fowls owned by the
well-endowed are equally significantly larger than those for the poorly-endowed.

Table 5: Selected households wealth indicators by wealth category in Zimbabwe
Wealth category
Poorly-endowed Well-endowed
(322) (275)
Economic index -0.79 0.92
Female headed households (%) 64.3 35.7
Male headed household (%) 48.2 51.8
Access to relief seed program (%)
Beneficiaries 54.4 45.6
Non-beneficiaries 52.4 47.6


Harrow


Ox-plough

Wheel barrow

Radio set

Bicycle

Scotch cart

Pair bullock


36

28


0 20


100 120 140


0 Poorly-endowed U Well-endowed
Figure 5: Proportional distribution of assets by wealth group












1000%


80% -

60% -

40% -

20%


Cultivated farm Cattle Small ruminants Fowls
Asset type

E Poorly-endowed U Well-endowed
Figure 6: Proportional distribution of productive assets by wealth group




Maize varietal knowledge

Farmers grow both maize hybrids and OPVs. Identifying individual varietal names is sometimes
a challenge to them. For instances, it is common to hear farmers mention the names of some
varieties by the name of the seed company that has the proprietary rights to the variety (e.g.,
PANNAR), or the symbolic names given by seed companies to describe the maturity group (e.g.,
imbizi, or zebra). Popular maize varieties among farmers include the SeedCo series, the ZMs and
Kalahari early pearl (KEP) (Table 6). Clearly, the dominant seed procured from the markets was
the SC series while the ZM series and KEP were mostly distributed by NGOs.

Data in Table 6 (column 6) and Table 7 suggest that some farmers recycle both maize hybrids
and OPVs. Asked if it was a common practice to recycle OPVs and hybrids, the majority of them
answered "yes" but about half of them actually recycle both. Most farmers think OPV seeds are
relatively cheaper than hybrids but agreed that they were hardly available on the local markets.
Seed venders are less willing to retail OPV seeds than hybrid seeds in the outlying areas because,
apart from the fact that farmers in these areas have low purchasing power, it is believed that they
would after purchasing once they would continue to recycle and not procure any fresh seed.
Because farmers in Seke had little experience with OPVs, none of them could compare the seed
costs but over 90% of them agree that hybrids are more readily available. One reason for
availability of hybrid seed in Seke is its proximity to Harare with a well-developed seed
infrastructure compared to the other districts. The results, however, show a split in the proportion
of beneficiaries who think that the yields of recycled OPVs and hybrids are similar.









Table 6: Sources of maize seed planted by farmers in Zimbabwe
Govern-
NGO ment Seed Neigh- Other
support support fairs bor Market Recycled sources Total
ZM 521 35.7 0.2 0.5 0.7 0.5 37.6
ZM 421/623 2.4 2.4
MATUBA 3.4 0.4 0.1 0.7 0.1 4.6
Kalahari early pearl 9.7 0.8 0.7 11.1
Hickory king 0.2 0.2 1.8 0.5 2.7
Other OPVs 0.1 0.3 0.0 0.4
SC 513 1.0 1.3 0.6 11.1 1.7 1.3 16.9
SC 401 0.4 0.2 0.1 0.1 2.5 0.4 0.4 4.1
SC 501 0.4 0.2 1.2 0.1 0.1 2.0
SC 517 0.1 0.1 0.8 0.1 1.0
SC 403 0.4 0.1 0.1 0.4 0.1 0.1 1.1
SC 407 0.2 0.1 0.4 0.1 0.1 0.9
Other SC series 0.8 0.1 0.1 1.1
REDCORK 0.1 0.1
R 215/201 0.2 0.2
PIONEER 1.2 0.3 1.1 0.2 2.8
PANNAR 3.6 1.4 4.7 0.1 0.5 10.4
Other hybrids 0.1 0.4 0.2 0.6
TOTAL 58.4 3.4 1.0 1.0 24.5 7.1 4.5 99.9


Comparing the physiological characteristics of OPVs and hybrids, fewer farmers in Makoni
compared with those in all other districts consider OPVs superior to hybrids. In particular, two-
thirds of them think OPVs are inferior to hybrids in terms of their resistance to field and storage
pests and tolerance to low soil fertility or drought (Table 7). More than three-quarters think
OPVs yield less and that the cobs are smaller than those of hybrids. In contrast, over 50% of the
farmers in the other districts think otherwise for most of the characteristics mentioned above,
except for tolerance to soil fertility, where farmers in Seke think OPVs are superior to hybrids. In
terms of maturity, over 70% of the farmers in all the districts are convinced that the available
OPVs mature earlier than hybrids, compared to about 40% of those in Makoni district.

Similarly, most farmers in Makoni district do not believe that sadza or roasted green maize from
OPVs are more palatable than those from hybrids, contrasting sharply with the perceptions of
farmers in all other districts. The reason for the strong dichotomy of perceptions about the
comparative characteristics of OPVs and hybrids between farmers in Makoni district and all
others interviewed is not clear.

















S00oo0000 (



100 "~ Tr -






00 00 (f 'OT





00 00I'()'




0o 00 m r- 00

00o00I'O






0 00 m(f


I 6 dr. CD C 0
o- r- I'D 00 ZTr W m ZT


\0 C C o W)








m 00 W) m m













6 r-- od I'D I'D o --
01) 0'"D t- m m--


-e



oa -


-ee


>) C







". -i 0 "0 0
Sa- -a a




>- ? -C S-o0 o)
>,OOo C
Off 4-, 4 O-c^


Q~)


o00












ooo


'c





t o
?a



1 ^






S
^a,




fi
I d





^Q


o *C



e -g



a- ^- ^


0 0O 00
~0000
aa 3S 0
-2 -2 ^
-^ ^ -

r^ r^ -<^ <-
a
-4 4- T T
Vl I f^ 'f
.^ .^ l) fl
VI V ^


-e
S-ef





CJC
NO













~NO



00-C0


e






0-e
a





-









00










Maize seed relief program activities


Maize seed distributed to beneficiaries

Maize seeds were distributed to beneficiary households in the Makoni, Hurungwe, Shurugwe,
Chivi, Seke, and Gwanda districts. Of the 597 respondents, 417 benefited from the PRP seed
relief program. An estimated 26% (or 108) of the 417 beneficiaries received maize seed in
2003/04, 52% (or 216) in 2004/05 and 76% (or 316) in the 2005/06 (Table 8). Across the 3
seasons, 54% of them received seed more than once.

Table 8: Information flow between NGOs and farmers
Hurun- Shuru- Chivi Seke Gwanda Whole
Makoni gwe gwe sample
--------------------------------------2003/04-----------------------


Number of relief seed
beneficiaries
Beneficiaries informed of type of
seed to be given (%)
Beneficiaries who thought the
seed was supplied on time (%)
At least one extension visit by
participating NGO (%)
Beneficiaries who received OPV
maize seed (%)

Number of relief seed
beneficiaries
Beneficiaries informed of type of
seed to be given (%)
Beneficiaries who thought the
seed was supplied on time (%)
At least one extension visit by
participating NGO (%)
Beneficiaries who received OPV
maize seed (%)

Number of relief seed
beneficiaries
Beneficiaries informed of type of
seed to be given (%)
Beneficiaries who thought the
seed was supplied on time (%)
At least one extension visit by
participating NGO (%)
Beneficiaries who received OPV
maize seed (%)


14 17 26 5 25 118


32 57 47 46

48 36 71 58


43 0.0


35 37


- 48 46

- 84 59

24 28


68 50 29 46 84 56
--------------------------------------- 2004/05------ ------------

50 45 42 49 8 39 233

40 69 50 47 50 62 53

32 33 45 47 63 74 49

44 18 36 41 38 21 33

74 98 60 49 75 97 75
--------------------------------------- 2005/06------ ------------

67 38 53 69 67 47 341

39 74 57 57 73 70 62

31 37 51 58 75 83 56


5 25 61 52 23 35


78 79 79 90 99 100









Beneficiaries in the Seke district appear least favored in terms of seed distribution in the first two
years of the program, notwithstanding the fact that a larger proportion of households there than
in the other districts are poorly-endowed.

In terms of type of seed distributed, Table 9 suggests that the ZM series (mainly ZM 421, 423,
523 and 521) were the most popular among the OPVs distributed, while among the hybrids, the
SC series (mainly SC 530, 513, 501, 413, 405, 403, and 401) dominated. Figure 9 shows that in
the three years of the program, relatively more poorly-endowed than well-endowed households
benefited, suggesting that the distribution was pro-poor at the district levels.

The quantities of seed received by household varied from 5 kg in Seke in the 2005/06 crop
season to 16 kg in Hurungwe in the 2003/04 crop season when the program started (Table 10).
Whereas over 70% of the farmers in Shurugwe and Gwanda thought that the seed was delivered
on time for planting in 2003/04 crop season, less than half of those in all other districts except
Seke (where the representative sample contained only five beneficiaries and was not included in
that analysis) agreed. The proportion of farmers, especially in Makoni, Shururgwe and Gwanda
districts, agreeing that seeds were distributed on time decreased over the years.

Table 9: Maize varieties distributed under the PRP program over the last three years (%)
Cropping year
2003/04 2004/05 2005/06
(n=118) (n=233) (n=341) Average
ZM series1 29.7 42.9 66.3 46.3
Kalahari early pearl 17.8 15.0 16.4 16.4
SC series2 22.0 9.9 3.8 11.9
Pannar 17.0 10.3 6.5 11.2
Matuba 8.5 16.7 5.0 10.1
Pioneer 1.7 3.4 1.2 2.1
Other hybrids3 2.5 1.7 0.9 1.7
Hickory king 0.9 0.0 0.0 0.3
Total 100.0 100.0 100.0 100.0
Note: 'ZM series include: ZM 421, ZM 423, ZM 523 and ZM 521;
2SC series include: SC530, SC 513, SC 501, SC 413, SC 405, SC 403, and SC 401;
3Other hybrids include: R 215, R 201 and Imbizi


Table 10: Average quantity of seed in (kg) received by beneficiary farmers
District 2003/04 2004/05 2005/06
Makoni 14.2 (6.2) 14.5 (5.4) 11.9(6.5)
Hurungwe 16.4 (12.2) 12.1(3.1) 12.1 (4.3)
Shurugwe 15.3 (11.1) 10.7(7.1) 13.0(9.2)
Chivi 8.8 (2.1) 10.7 (5.3) 10.4 (9.9)
Seke 13.0(6.7) 11.1 (6.0) 5.4 (2.3)
Gwanda 12.0 (7.8) 10.8 (7.7) 10.7 (7.3)
Note: In parentheses are standard deviations










200
180
S160 -
140
120
100 -
80 -

40
20

202003 014 2(004 05 2005/06 200-3 04 2(004 05 2005/06

Any maize seed OPV maize seed
Year and seed type

Poorly-endowed U Well-endowed
Figure 7: Distribution of relief seed type by wealth group




Effectiveness of maize seed assistance to vulnerable
households

As noted earlier, the decision to supply farmers with OPVs was to give them the possibility to
recycle without severe loss in genetic vigor of the seed. Under the PRP program, NGOs were
expected to inform farmers of the types of seed being distributed and the need to select, store and
re-use the seed the following and subsequent seasons. Community level spillovers are expected if
farmers share the selected seeds with their neighbors. To assess the effectiveness of the relief
program, the following operational definitions were adopted:

1) The flow of information from NGOs to beneficiary households on the types of maize
seed distributed and the need to select the OPV maize seed, and
2) The proportion of farmers who actually selected OPV maize seed and shared with their
neighbors for re-use in subsequent years.

Less than half of the beneficiaries in the first year of the program were informed of the type of
seeds to be provided but the proportion (and hence number of beneficiaries) increased to more
than 60% over time (Figure 10). Information on OPVs was limited to the fact that they can be
recycled. Compared with farmers who were informed of the type of seeds to be distributed,
fewer farmers were paid extension visits by the participating NGO during the crop growing
season. In terms of farmers' preferences for the seed distributed, far fewer OPV than hybrid
maize variety recipients liked the varieties possibly due to the fact that few farmers participated
in the choice of variety to be distributed (Figure 11).












An estimated 408 of the 417 sampled beneficiaries ever received OPV maize seed in one of the
three years of the program, 65% receiving more than once. In the first year, 118 (or 29%)
received, in the second 233 (or 57%) and in the third, 341 (84%).


40

- 30

20

10


0
0


Informed of variety Timely seed supply At least 1 field visit Availability of seed
type byNGO onmkt

U 2003/04 (n=118) U 2004/05 (n=233) 0 2005/06 (n=341)

Figure 8: Access to and use of information by seed relief beneficiaries


90
80
70
60
50
40
30
20
10
0


Hybrid


I


OPV


Preference for variety


Participated in choice
of variety


2003/04 0 2003/05 0 2003/06
Figure 9: Information on seed type and beneficiaries' preferences for the seed


I









Whereas all the OPV beneficiaries in Shurugwe and over 50% of those in Makoni, Hurungwe
and Chivi districts were informed about selecting seed for replanting, less than half of those in
Gwanda benefited from such education (Table 11). As indicated in Table 11 and Figure 12, the
proportion of OPV maize seed beneficiaries informed about selecting seed increased from 55%
of the 118 (or 65 beneficiaries) in the first year to 62% of the 341 (or 211 beneficiaries) in the
third. In terms of provision of seed selection information, Figure 12 shows that 25% received
such information more than once, and 75% only once. In terms of teaching farmers how to select
or store seed, less than 50% of them benefited in the three consecutive years (Figure 13 and
Table 11).

To have an idea about household preferences for the distributed OPVs, beneficiaries were asked
if they would have bought similar varieties from the market using their own money. More than
half of them answered "yes," suggesting that the varieties distributed met their preferences.
However, the proportion liking the varieties decreased over time, possibly because farmers were
not consulted on the types of seed they needed. The unavailability of the varieties on the local
markets was also of concern to more than 80% of beneficiaries (Table 11).

Nearly all beneficiaries were happy with the taste of the "sadza" made from the distributed seeds.
Notwithstanding the fact that the varieties seemed to meet the preferences of most beneficiaries
in terms of performance in the field and food qualities, less than 50% of them actually selected
seed. In fact, a far smaller proportion of those informed about selecting the seed actually did. In
terms of actual number of farmers selecting seed, 24 did in the first year, 50 in the second and
126 in the third. Whether or not information on seed selection coupled with teaching farmers
how to select had a greater impact on seed selection was assessed. Figure 14 shows that the
probability of getting a farmer to select seed is much higher if the farmer is informed and taught
how to select. This means that greater benefits would be achieved by complementing the
information with field level teaching on best practices in seed selection.

Although fewer well-endowed households were given OPV maize seed, they were favored in the
dissemination of seed selection information (Figure 15), possibly explaining why more of them
selected seeds than the poorly-endowed. Gender bias in terms of the information flow was
somewhat mixed. The proportion of males that got seed selection information in the first year of
the program was greater than that for females, and remained relatively constant over time, while
that for females increased linearly and surpassed males in the third year (Figure 16). In terms of
seed selection per se, the same Figure shows that females appear more willing than males.

Table 12 suggests that nearly all farmers who select seeds re-plant. But sharing the selected seeds
with neighbors seems unpopular among them. In the second year of the program, beneficiaries in
Shurugwe were more willing to share seeds with their peers, contrasting the actions of their
colleagues in Gwanda.












W0 m N00\-0 0o
Sinv~t- mMoooIcc





' m CN CN ZT I'D C 00


So ct- oo m ZT c






0 W I'D 0 ID "OO (N
m W) N N 00 0')


tf) 0tO) oOC I'












c M ooooi iD
O'N m- i0


























m ZT m N ZT r- -
OOmmi^Mmt\O







(N (N ( ^ ^ oO(-n (-
vi^ ^t mM ^- t '


( 0n ZT 0 N


0



00 30
O









CA CA ,0
000 c
* ) a -



fJ.-00


000 00

c^ ^ ^ ^ ^
i^ ^ ^ i i
a a a a alr F
iu o o < <
ra a a r r


0



0^



3
OO -





m -
Ca






00

C






0 0
C41












0 j
c^c^

iu iu


0


0



0



-e








~_$
-e a





cn~n



0000
QCccjQ

C i


^00

. '0



a .

.u iu .u .u

a a a a~
iu u i i
ra ra raraF


0
0

-) 0










Can







000
./j CA '


A 8
S"


000





000
^3 *S ^

-g-S
^ ^3
'> 'Z CO
0 0 0~
.a a .
r^ ^ rF
CO O C
iU U i
*G *G *













60

50

S40

30

20

10

0
Advised to select Taught to select Taught to stor

U 2003/04 (n=67) U 2004/05 (n=174) O 2005/06 (n=299)
Figure 10: Flow and use of OPV seed selection information by beneficiaries


e


Figure 11: Access to recycling information by OPV maize seed beneficiaries


Thrice
Twice 5%
200 oaL 7


Once
75%











70

60

50

40

30

20

10



2003/04 2004/05 2005/06

Year

U Advised U Taught

Figure 12: Seed selection behavior of beneficiaries based on information and knowledge


2003/04 2004/05 2005/06 2003/04 2004/05 2005/06
(n=32; 35) (n=89; 85) (n=163; 136) (n=32; 35) (n=89; 85) (n=163; 136)
/
BendfiSries given seed selection information Beneficiaries who selected seed
(Note: The first "n" is the sample Year and activity
size for the poorly-endowed and
the second for the well-endowed) U Poorly-endowed U Well-endowed

Figure 13: Information on and seed selection behavior of beneficiaries by wealth group
























__T_


In


In


2003/04 2003/05 2003/06
(n=30; 37) (n=76; 98) (n= 122; 177)
A t
Benefilaries given seed selection information
I,


In


2003/04
(n=30; 37)


2003/05
(n=76; 98)


2003/06
(n= 122; 177)


Beneficiaries who selected seed


Year and activity

* Female 0 Male


Figure 14: Information on and seed selection behavior of beneficiaries by gender


Table 12: Distribution of selected seed within the communities1
Makoni Hurungwe Shurugwe Chivi Gwanda
Beneficiaries who re-planted
selected (%) 67 88 86 84 91
Beneficiaries who donated
selected seed to neighbors (%) 19 29 55 31 11
Non-beneficiaries who
received seed from
beneficiaries (%) 11 20 12 0 20
Note: 1Seke was not included because the sample size was either zero or one


Factors determining the recycling of relief seed by
households

The decision by relief seed beneficiaries to recycle seed or not was influenced by factors such as
the information they had and past experience. To make realistic recommendations on how to
improve recycling of OPV among farmers in selected communities, collective and relative
impacts of selected factors were examined using a simple Probit4 model. The model measures


4 A mathematical treatment of the Probit model is not included in this paper as its usage is common in applied economics
research. Thorough treatments of the model may be found in Greene (2000) or any standard econometric book.


60

50-
0
| 40 -

30-

^ 20-
-


In


(Note: the first "n" is the
sample size for females
and the second for males)









the probability that a seed relief beneficiary would recycle OPV maize seed received. The
following three hypotheses were tested:

Farmers would not recycle OPV seed even if provided seed selection information,
Previous knowledge on recycling OPV maize has no influence on seed selection
decisions by beneficiaries, and
Maize varietal preferences have no influence on farmers' seed selection behavior.

The probit model was run on a sample size of 408 OPV beneficiaries. At the 99% confidence
level, the results5 rejected the null hypothesis that, even if provided information on seed
selection, farmers would not select, in favor of the alternative hypothesis that, given the
information, farmers would select. That is, beneficiaries would consider seed recycling a suitable
option for securing seed for subsequent seasons. The marginal effects suggest that, with 99%
confidence, the probability of getting a farmer to recycle will increase by 22%, if the NGO
representative providing the seed informs him/her about doing so.

The null hypothesis of no influence of previous experience on seed selection was rejected in
favor of the alternative that previous experience has a significant effect on seed selection
decisions by farmers. The probability that a farmer initially without any experience with OPV
maize seed recycling would recycle will increase by 24% with just a year's experience with the
seed. This seems to confirm the proportionate increase in recycling over time, especially among
those who had previously handled OPVs under the program. With time, beneficiaries developed
the confidence to recycle and more readily did it than when initially exposed to the practice.

If provided with seed that met their preferences, beneficiaries were more likely to select than
otherwise. This supports the alternative hypothesis that preferences have a positive impact on
seed selection decisions. That is, by targeting varieties farmers themselves choose, the
probability of increasing the recycling rates among them would increase by 13%.

Apart from Chivi district where beneficiaries appear more willing to select seed than in Gwanda
(the base district), no significant difference was observed among all the districts. There is a 33%
greater probability of convincing beneficiaries to recycle in Chivi than in Gwanda; in other
words, more effort would be needed to convince farmers in Gwanda and all other districts to
recycle seed.

The effectiveness of getting beneficiaries to select seed varied with the participating NGO.
Compared with CAFOD (the base NGO)6 and all other NGOs, World Vision International,
Action Aid and Christian Care appear to have made significant positive impacts on their
beneficiaries' decisions to select seed.





5 The estimated regression results can be obtained directly from the lead author.
6 CAFORD was deliberately chosen simply because it supplied the largest quantity of seed to beneficiaries in the
target districts in 2006, but any NGO could have been chosen without affecting the results.









Lessons learned from maize seed relief to vulnerable groups
in Zimbabwe

The flow of information from NGOs to program beneficiaries on seed selection

Survey results clearly show that, over the three-year (2003/04 2005/06) period when OPV
maize seeds were distributed to vulnerable farm households under the PRP, decisions regarding
the type of seed to be distributed were mostly at the discretion of the participating NGO. Less
than 10% of the beneficiaries noted ever participating in decision-making. About 20% of them
were informed beforehand regarding the type of seed to be provided. Seeds distributed
(especially the OPVs) were not commonly found on the market, mainly because of seed
companies' reluctance to produce and sell OPV maize seeds. Recent reports suggest that seed
companies produce OPV maize seeds on request through tenders by NGOs. Companies believe
that if OPV maize seeds are put on the market, farmers may buy once and continue to recycle,
thereby not guaranteeing them (companies) a stable market. However, the current results showed
that farmers recycle both OPV and hybrid maize and are split on whether or not recycled OPVs
are superior to recycled hybrids in yield.

In the first year of the program, about half the beneficiaries were informed about the need to
select, store, and re-use OPV seed by the participating NGOs. By the third year, this proportion
increased to 62%. The proportion taught how to select was far less than that simply advised to
select. Disaggregating the data by gender, the results show that more males (60%) than females
(50%) were informed about selecting in the first year. However, the proportion of females
advised about selecting increased to 68% by the third year, surpassing that for males, which
remained relatively constant. When disaggregated by wealth group, relatively more well-
endowed than poorly-endowed households were advised to select over the three years, although
more of the latter were given OPV maize seed.

The relative importance of factors influencing farmers' seed selection decisions

Not all beneficiaries advised about and/or taught to select, store, or re-use seed complied. The
survey results suggest that each year, fewer than half of those advised about selecting actually
selected. If seed selection information is complemented by teaching farmers how to select, the
proportion of those selecting increases. Unfortunately, far fewer beneficiaries were actually
taught how to select. Aggregated over the years, the flow of seed selection information was
gender insensitive. However, females appear more aggressive in selecting seed. Disaggregating
the data by wealth category, fewer poorly-endowed than well-endowed farmers actually selected
seed, reflecting the fact that the flow of information was skewed against them.

Some important conclusions could be drawn regarding factors important in determining
beneficiaries' seed selection decisions. Based on the tobit regression analysis (not reported
here)7, the survey results demonstrate that if given seed selection information, the probability
that farmers would select seed will increase significantly. As expected, the probability that a
beneficiary would select seed will increase significantly if the seeds donated meet his/her

7 Contact the lead author if you need the detailed results.









preferences. Farmers with previous knowledge on OPV maize seed selection show a greater
propensity to select than those experiencing it for the first time. Some NGOs appear more
effective than others in convincing their beneficiaries to select seeds.


Concluding remarks

Based on the findings of the study the following recommendations are made to improve seed
assistance to vulnerable groups in Zimbabwe to ensure greater spillover effects.

Experience with OPV maize seed recycling is important in influencing farmers' recycling
decisions. Ways to extend similar interventions for a few more years should be explored.
In this way, many more farmers would have experienced handling OPV maize seed and
hence be recycling seed, ensuring greater spillover effects within the communities.
Given that the relatively "well-endowed" farmers are more willing to recycle OPV seed,
targeting them can potentially ensure large-scale spillovers.
NGOs should endeavor to ensure that all beneficiaries be provided with seed selection
information. They should follow up the information with field-level
training/demonstration of seed selection practices during the crop growth period. This
could help foster greater adoption of selection than simply providing information.
Seed selection information should emphasize the need to select only OPV maize, since
farmers are inclined to select any type of variety.
Participating NGOs should share their seed selection information strategies so that less
successful ones can learn from those able to ensure that large proportions of their
beneficiaries select seed.
Beneficiaries should be given the opportunity to select the varieties they want, because
the probability of recycling is higher if a variety meets their preferences.
Seed companies should promote OPVs: farmers recycle both OPVs and hybrids but, if
given a choice, they will purchase fresh seed every year, thereby creating an OPV
market.
A simple varietal naming system should be developed to help farmers remember the
names of the varieties they receive and to distinguish between hybrids and OPVs.










Annex 1:
Coverage of past OPV maize seed distribution programs
Province District Ward Participating No. No. Year Seed
NGO assisted selected received
(kg)


Masvingo
Masvingo
Masvingo


Masvingo
Masvingo
Masvingo

Masvingo
Masvingo
Masvingo

Matebeleland South
Matebeleland South

Matebeleland South
Matebeleland South
Matebeleland South
Matebeleland South

Matebeleland South
Matebeleland South
Matebeleland South
Mashonaland West
Mashonaland West
Mashonaland West
Mashonaland West
Manicaland
Manicaland
Manicaland
Manicaland

Manicaland
Manicaland
Manicaland

Manicaland


Chivi 1 CARE
Chivi 1 IFRC
Chivi 1 Danish Red
Cross
Chivi 4 CARE
Chivi 4 IFRC
Chivi 4 Danish Red
Cross
Chivi 10 IFRC
Chivi 10 CAFOD
Chivi 10 Danish Red
Cross
Gwanda 2 IFRC
Gwanda 2 Danish Red
Cross
Gwanda 2 HELP
Gwanda 4 IFRC
Gwanda 4 WVI
Gwanda 4 Danish Red
Cross
Gwanda 4 WVI
Gwanda 14 WVI
Gwanda 14 WVI
Hurungwe 10 Cadec
Hurungwe 10 ChristianCare
Hurungwe 14 GOAL
Hurungwe 17 ChristianCare
Makoni 3 ZimPro
Makoni 3 MDA
Makoni 15 ZimPro
Makoni 15 FACT
Rusape
Makoni 15 GOAL
Makoni 15 MDA
Makoni 19 FACT
Rusape
Makoni 19 MDA


425
390
390


451
390
338

79
996
79

54
54

1,000
54
1,174
54

1,448
745
1,024
2,193
1,978
500
1,971
1,437
1,000
445
360

670
400
360

348


2005/06 2.5
2005/06 10
2004/05 10

2005/06 2.5
2005/06 10
2004/05 10

2005/06 10
2004/05 10
2004/05 10

2005/06 10
2004/05 10

2004/05 8
2005/06 10
2005/06 5
2004/05 10

2004/05 6
2005/06 5
2004/05 6
2005/06 10
2004/05 5
2004/05 5
2004/05 5
2005/06 10
2004/05 15
2005/06 10
2005/06 5

2004/05 5
2004/05 15
2005/06 5

2004/05 15










Annex 1: (Cont.)
Province District Ward Participating No. No. Year Seed
NGO assisted selected received
(kg)
Mashonaland East Seke 1 Oxfam 625 17 2005/06 5
America
Mashonaland East Seke 1 CAFOD 223 6 2004/05 10
Mashonaland East Seke 5 Oxfam 625 17 2005/06 5
America
Mashonaland East Seke 5 CAFOD 223 6 2004/05 10
Mashonaland East Seke 8 Oxfam 625 17 2005/06 5
America
Mashonaland East Seke 8 CAFOD 223 6 2004/05 10
Midlands Shurugwi 5 HELP- 100 4 2005/06 10
GERMANY
Midlands Shurugwi 5 IFRC 11 0 2005/06 10
Midlands Shurugwi 5 IFRC 12 0 2004/05 20
Midlands Shurugwi 5 Christian 301 11 2004/05 15
Care
Midlands Shurugwi 10 HELP- 100 4 2005/06 10
GERMANY
Midlands Shurugwi 10 IFRC 11 0 2005/06 10
Midlands Shurugwi 10 IFRC 102 4 2004/05 20
Midlands Shurugwi 10 Christian 424 16 2004/05 15
Care
Midlands Shurugwi 15 HELP- 100 4 2005/06 10
GERMANY
Midlands Shurugwi 15 Action Aid 620 23 2005/06 4
Midlands Shurugwi 15 MASO 100 4 2004/05 3
Source: FAO-Zimbabwe Office, 2006







Annex 2:
Distribution of cultivated land to crops in selected districts in Zimbabwe








.- 1._ L!.1'-


Source: File survey data, 2006


\\


2%~4 :


. .-N .


d--


1


:- r






ISBN: 970-648-148-6


1ICIMMYTR
International Maize and Wheat Improvement Center
Apdo. Postal 6-641, 06600 Mexico, D.E, Mexico
www.cimmyt.org




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