&+$57,1*$&2856(
72)22'6(&85,7<
)256287+(513529,1&(
MID-TERM EVALUATION OF THE
CARE LIVINGSTONE FOOD SECURITY
PROJECT
Associates in Rural Development, Inc.
For USAID Zambia
Contract PCE-I-00-99-00001-00
5 June 2000
 -RAISE
CHARTING A COURSE TO FOOD
SECURITY FOR SOUTHERN PROVINCE:
MID-TERM EVALUATION OF THE
CARE LIVINGSTONE FOOD SECURITY
PROJECT
by
Andrew Lyons
Anthony Johnson
Godfrey Mwila
Austin Muyaba
Presented to
USAID/Zambia
under the Rural and Agricultural Incomes with a Sustainable Environment (RAISE)
Contract No. PCE-I-00-99-00001-00: Task Order 11
June 2000
ARD-RAISE Consortium
1601 North Kent St., Suite 800
Arlington, VA 22209
Tel: 703-807-5700, Fax: 703-807-0889
gkerr@arddc.com
The ARD-RAISE Consortium:
Associates in Rural Development, Cargill Technical Services,
Associates for International Resources and Development, CARANA Corporation,
A.P. Gross & Company, International Programs Consortium, Land O’ Lakes,
Purdue University, and TechnoServe
This report is available from the Development Experience Clearinghouse,
http://www.dec.org
LFSP Mid-Term Evaluation ARD
i
EXECUTIVE SUMMARY
The CARE Livingstone Food Security Project (LFSP) Phase II is currently in its
fourth year under a five year financing agreement with USAID. It evolved from a
food relief project that was established in response to repeated food shortages in the
early 1990s in Zambia's Southern Province. These food shortages were caused
primarily by drought, but were compounded by a changing agribusiness environment,
long-term climatic shift, and the cumulative loss of production assets for smallholder
farmers.
The primary objective for this midterm evaluation is to determine whether USAID
investments in LFSP are achieving their desired impact, and explain why or why not.
A second objective is to generate ideas on how the impact of USAID investments can
be improved, and a final objective is to generate ideas on how LFSP experiences can
influence ongoing or future USAID and other institution investments in increasing
rural incomes and improving food security.
LFSP currently falls under USAID's Strategic Objective One: "Increasing the rural
incomes of selected groups". Over the past four years, the program has made
important contributions towards SO1 by increasing agricultural production in
beneficiary households through the use of appropriate technologies developed within
the project agroecological context. This is quite a notable achievement considering
that the service area is in the drought-prone districts of Kazungula, Kolomo, and
Livingstone, which have a recent history of food relief.
The overall goal of the project is to improve food security and vulnerability to drought
by addressing the underlying causes of food insecurity. The project operates through
four self-reinforcing lines of action:
1. building the capacity of community-based organizations (CBOs)
2. developing improved farming system practices
3. expanding the adoption and effectiveness of water harvesting technologies
4. supporting small scale income generation
The evaluation team finds that LFSP has overall made significant progress in
achieving its goals in the first three objectives. The development of community-based
organizations (CBOs) provided the means to identify the underlying threats to food
security and develop interventions which are demand-driven and relevant to
beneficiary needs. The strong focus on CBO capacity building has also proven to be
an effective way of increasing the efficiency of delivery of services such as seed
inputs and extension training, and strengthened the ability of rural people to manage
their own development activities so they may gradually become less dependent on
LFSP. The benefits of working together in small groups have been solidly
demonstrated to both rural farmers and other institutions; this may very well be the
most lasting legacy of LFSP.
The introduction of early maturing drought resistant crop varieties may be the
intervention which has had the greatest impact on improved food security for the
LFSP Mid-Term Evaluation ARD
ii
estimated 26,000 direct beneficiaries and their families. These improved varieties are
bulked in a community-based seed multiplication scheme, ensuring sustained supplies
of seed stocks. Although yield data is somewhat inconclusive, project monitoring data
shows clearly that there is greater food availability in the communities than there was
before the project started.
The improved farming practices are complimented by a program to improve water
supply. Over 20,000 farmers and their families benefit from the 80 water harvesting
structures that have been built or rehabilitated with USAID financing. Although the
water harvesting structures have not made much impact on crop production, which
remains largely rainfed, they have had a demonstrated positive impact on the health of
people. Improved water supply has also benefited the production of livestock and
vegetables which are sources of both food and income.
However LFSP has been challenged to demonstrate significant impact in the last
objective, income generation. The project has found it difficult to boost production of
cash crops and other goods and services to a volume and level of efficiency high
enough to compete in the market economy. It was also sidetracked on a savings and
credit scheme that was based on an urban model and never achieved success in small
scale rural communities. However the project has demonstrated on a limited scale a
number of approaches for developing viable group enterprises, and the recent
reorientation on strengthening market linkages bodes well for future progress.
The overall strategy of LFSP is to demonstrate how new technologies, community
structures, and participatory methods can improve food security and increase rural
income, and then hand-over the lessons learned to institutions and investors with
greater longevity and resources, such as the private sector, government agencies, and
the CBOs themselves. Thus as the project nears the end of the current funding period,
its major challenge is to begin to gradually move away from direct facilitation of
activities, and develop institutional partnerships or structures that will carry the
proven methods beyond the project completion date. The strong focus on action
research has made laudable strides in extracting lessons learned from technology
demonstrations, however LSFP has had much less success in building the necessary
bridges with private sector and government institutions that will be needed to sustain
impact. The project acknowledges the need to strengthen partnerships and improve
market linkages, but may have to reexamine its current staffing skills and activity
focus to prepare for the transition period.
The evaluation team recommends that USAID continue to support LFSP, and
encourage the recent programmatic reorientation on developing stronger market
linkages, NRM, and livestock health. The people of Kalomo, Kazungula, and
Livingstone Districts will continue to benefit from these efforts, and there is great
promise that the lessons learned will expand to other sectors and areas of Zambia.
LFSP Mid-Term Evaluation ARD
iii
ACKNOWLEDGEMENTS
This mid-term evaluation could not have been possible without the generous
assistance of the staff of the LFSP and CARE International. Many thanks go to the
project team in Livingstone for putting up with not just one but two consultants on
two separate occasions for a total of three weeks of fieldwork. In particular thanks go
to Sylvester Kalonge and Robby Mwiinga for generously offering their time to the
consultants, and Florence Chawelwa, Agnes Bwalya, and Yvonne Siankweleku for
ensuring that the logistics of the evaluation went smoothly. Catherine Pongolani,
Fungaloko Patson, Felix Lombe accompanied the consulting team in the field for
several days at a time, and provided valuable technical information. Jasper Hatwiinda
was invaluable in helping the evaluation team navigate through a mountain of
monitoring data, and worked extra hours to enter data from the Community Self￾Monitoring ledgers, even while suffering from malaria. Both Sarah Mwale and Alice
Whitehead went above and beyond the call of duty by taking a break from their
annual leave to plot the location of AMCs and enter CSM data respectively. Fred
Sichone put together an excellent budgetary analysis of the project expenses tailored
to the evaluation scope of work. Special thanks are also due to the Permanent
Secretary of Southern Province, Mr. D.H. Hakayobe for sharing his time.
In the CARE Lusaka office, thanks go in particular to Mark VanderVort and Godrey
Miiti, who provided background information and documentation and liased with the
Livingstone office to facilitate the evaluation.
Many others who were instrumental in providing information for the evaluation are
listed in Annex I. Anything positive that comes out of the evaluation deserves to be
shared with the many people who made this report possible. However responsibility
for any errors or misrepresentation of the project lies only with the consultants.
LFSP Mid-Term Evaluation ARD
iv
TABLE OF CONTENTS
1.0 Objective.............................................................................................................1
2.0 Methodology.......................................................................................................1
3.0 Background ........................................................................................................2
3.1 Prelude to a food security crisis ........................................................................2
3.2 The Physical Environment................................................................................3
3.3 Rainfall.............................................................................................................4
3.4 The People .......................................................................................................5
3.5 Land Tenure .....................................................................................................5
3.6 Agriculture .......................................................................................................6
3.7 LFSP Phase I, and the South-Western Drought Relief and Mitigation Project...9
3.8 Livingstone Food Security Project Phase II ....................................................10
4.0 Project Implementation ...................................................................................11
4.1 Project Components........................................................................................11
4.2 Project Resources ...........................................................................................21
4.3 Financial.........................................................................................................23
5.0 Beneficiaries......................................................................................................27
6.0 Performance Indicators....................................................................................32
6.1 Community institution and capacity building..................................................32
6.2 Appropriate and sustainable farming systems .................................................45
6.3 Improved water harvesting and natural resource management practices ..........52
6.4 Increased incomes and income earning opportunities......................................58
6.5 Reduced vulnerability to drought and improved household food security........67
7.0 Gender ..............................................................................................................70
8.0 Institutional Relationships ...............................................................................74
8.1 Ministry of Agriculture, Food and Fisheries (MAFF)......................................74
8.2 Other Government Institutions and Agencies..................................................75
8.3 Traditional Leaders.........................................................................................76
8.4 Private Sector .................................................................................................76
8.5 Other NGOs...................................................................................................77
8.6 SWOT Analysis of the Major Partnerships .....................................................77
9.0 Policy Issues......................................................................................................79
9.1 Agriculture Policy Context .............................................................................79
9.2 Forestry Policy Context ..................................................................................79
9.3 Impact of Government Policy on LFSP ..........................................................80
9.4 Impact of LFSP on Policy...............................................................................81
10.0 Other Issues ....................................................................................................84
10.1 Extension......................................................................................................84
LFSP Mid-Term Evaluation ARD
v
10.2 Seed Scheme ................................................................................................87
10.3 Monitoring and Research Issues ...................................................................91
10.4 Scaling up Technology .................................................................................99
10.5 Life After LFSP..........................................................................................101
11.0 Lessons Learned ...........................................................................................111
11.1 LFSP Lessons Learned ...............................................................................111
11.2 Expanding Lessons Learned .......................................................................114
12.0 Additional Research Needs ..........................................................................118
12.1 Return on the donor investment ..................................................................118
12.2 Project impact on household asset base.......................................................118
12.3 Institutional capacity of partner institutions ................................................118
12.4 Accuracy of Community Self-Monitoring Data ..........................................118
12.5 Impact of LFSP on Food Relief ..................................................................119
12.6 Impact of Technologies on Production........................................................119
13.0 Recommendations ........................................................................................121
13.1 Recommendations USAID's Future Investment Options .............................121
13.2 Programmatic Recommendations................................................................125
Annex I - People Contacted .................................................................................127
Annex II - Literature Consulted..........................................................................130
Annex III - Envisioning the future ......................................................................132
Annex IV - LSFP Conceptual Framework..........................................................134
Annex V - Technology Matrix .............................................................................135
Annex VI - Improved Seed Varieties Characteristics and Sources....................143
Annex VII - Scope of Work .................................................................................147
TABLE OF FIGURES
Figure 1. Kazungula, Livingstone, and Kalomo Districts in Southern Province,
Zambia.......................................................................................................................3
Figure 2. Digital elevation model of Kalomo, Kazungula, and Livingstone Districts. .3
Figure 3. LFSP Area Management Committees grouped by agroecological zone........4
Figure 4. 70 year annual rainfall from Livingstone Meteorological Station.................5
Figure 5. Demonstration of ripping using draught power in Siamasimbi AMC. ..........6
Figure 6. Worksheet from Katapazi PRA, 1995..........................................................7
LFSP Mid-Term Evaluation ARD
vi
Figure 7. 1995-1999 Seasonal rainfall from Livingstone Meteorological Station. .......8
Figure 8. Timeline of donor support for CARE's food security interventions in
Southern Province......................................................................................................9
Figure 9. Women engaged in PRA exercise..............................................................10
Figure 10. Interest Group receiving training on mungongo nut oil extraction............12
Figure 11. LFSP CBO Structures .............................................................................13
Figure 12. LFSP Area Management Committee locations.........................................14
Figure 13. Seed distribution......................................................................................17
Figure 14. Seed scheme repayment ..........................................................................18
Figure 15. Demonstration of intercropping maize with squash to conservation soil
moisture...................................................................................................................19
Figure 16. Vegetable garden at Mandia ....................................................................21
Figure 17. Organizational Chart of LFSP. ................................................................22
Figure 18. Timeline of vehicle procurement in LFSP. ..............................................23
Figure 19. Total expenditures (actual and projected) by category for LFSP 1996-2001.
................................................................................................................................24
Figure 20. LFSP total funds committed and expenditures by year.............................25
Figure 21. Total projected expenditures and actuals by budgeted category. ..............25
Figure 22. LFSP expenditures within Zambia, 1996-2000. .......................................26
Figure 23. LFSP overall cost per beneficiary. ...........................................................27
Figure 24. Household size distribution of project beneficiaries, 1998. ......................28
Figure 25. Average area cultivated per household for all crops combined.................28
Figure 26. Relationship between family size and area under cultivation, 1998..........28
Figure 27. Age and gender profile of LFSP households, 1998. .................................29
Figure 28. Wealth profile of households in LFSP, 1998............................................29
Figure 29. Number of direct beneficiaries by year ....................................................30
Figure 30. Location of new AMCs. ..........................................................................31
Figure 31. Dundumwezi PRA, 1998.........................................................................31
LFSP Mid-Term Evaluation ARD
vii
Figure 32. LFSP AMCs per year. .............................................................................33
Figure 33. Percentage of AMCs using Community Self-Monitoring Ledgers............37
Figure 34. Amount of data currently entered into the Community Self-Monitoring
database. ..................................................................................................................38
Figure 35. Food calendar from Makunka PRA, 1995................................................39
Figure 36. Quantities of seed distributed by AMCs and LFSP ..................................42
Figure 37. Training on constructing seed storage bins. .............................................43
Figure 38. Management of water points....................................................................44
Figure 39. Maize field of drought tolerant MMV400, Makunka, 1996.....................45
Figure 40. Average amount of seed planted per household by seed source................47
Figure 41. Percent of AMCs with demonstration farming systems projects...............48
Figure 42. Number of facilitators, actual and target. .................................................49
Figure 43. Number of crops and varieties distributed by LFSP. ................................50
Figure 44. Shifting kraal, Delevu..............................................................................51
Figure 45. Location of dams built in Kazungula since 1995......................................53
Figure 46. Siabandwa dam undergoing rehabilitation with tractor and dam scoop. ...53
Figure 47. Weir rehabilitation in Nyawa...................................................................54
Figure 48. Woman demonstrating the treadle pump for vegetable irrigation .............55
Figure 49. A water committee visits Siabandwa dam................................................57
Figure 50. Interest groups by category......................................................................59
Figure 51. Thatching grass ready to be sold..............................................................60
Figure 52. Locations of outgrower schemes..............................................................62
Figure 53. Household food sources, 1998.................................................................67
Figure 54. Average period with food for maize. .......................................................68
Figure 55. Post-harvest period with food, maize. ......................................................68
Figure 56. Average value of all household assets......................................................69
Figure 57. Average value of all production assets minus cattle. ................................70
LFSP Mid-Term Evaluation ARD
viii
Figure 58. Average total value of assets for the same set of households....................70
Figure 59. Gender representation in CBO general membership.................................71
Figure 60 - Gender representation in CBO leadership...............................................72
Figure 61. Proportion of Men and Women Cultivating Different Crops in 1998. ......73
Figure 62. Chief Sekute plants a guava seedling donated by the Forestry Department
................................................................................................................................75
Figure 63. To strengthen its institutional relationships, LFSP has recently conducted a
series of two partnership workshops, with a third one scheduled for June 2000. .......78
Figure 64. Seed repayment requires some quality control measures..........................88
Figure 65. Local seed banks ensure that seed is available for planting ......................90
Figure 66. Beneficiaries per VMC histogram ...........................................................94
Figure 67. A framework for expanding lessons learned. .........................................116
Figure 68. LFSP planning workshop ......................................................................117
LFSP Mid-Term Evaluation ARD
ix
ACRONYMS
ADMADE Administrative Management Design
ADO Agricultural Development Office
AMC Area Management Committee
ASIP Agricultural Sector Investment Program
CBO Community Based Organization
CIDA Canadian International Development Agency
CLUSA Cooperative League of the USA
CSM Community Self Monitoring
DACO District Agriculture Coordinator
DDCC District Development Coordinating Committee
DFID Department for International Development (UK)
FEWS Famine Early Warning System
FPST Food Production Trends Survey
GART Golden Valley Agricultural Research Trust
GRZ Government of the Republic of Zambia
IDE International Development Enterprises
IGA Income Generating Activity
LFSP Livingstone Food Security Project
M&E Monitoring and Evaluation
MAFF Ministry of Agriculture Food and Fisheries
MENR Ministry of the Environment and Natural Resource
MOU Memorandum of Understanding
NEAP Natural Environmental Action Plan
NRM Natural Resources Management
PRA Participatory Rural Appraisal
RIF Rural Investment Fund
RGBP Rural Group Business Program
SC Sector Coordinator
SCCI Seed Certification and Control Institute
SEAD Small Economic Activity Development
SWDRP South Western Drought Relief and Mitigation Project
T&V Training and Visit
UNZA University of Zambia
USAID United States Agency for International Development
USGS United States Geological Survey
VMC Village Management Committee
WASHE Water Sanitation and Health Education
ZFAP Zambia Forestry Action Plan
LFSP Mid-Term Evaluation ARD
1
1.0 OBJECTIVE
The primary objective of the CARE/LFSP midterm evaluation is to determine
whether USAID investments are achieving their desired impact, and explain why or
why not. To guide this assessment, the evaluation examines the performance
indicators for LFSP as set out in the 1996 Cooperative Agreement between USAID
and CARE International.
A second objective is to generate ideas on how the impact of USAID investments in
CARE/LFSP activities can be improved.
A final objective is to generate ideas on how CARE/LFSP experiences can influence
ongoing or future USAID and other institution investments in increasing rural
incomes, improving food security or managing natural resources.
2.0 METHODOLOGY
The evaluation took place over an eight week period between 2 April and
5 June 2000. For the first five weeks, the evaluation team consisted of three technical
specialists in the areas of agronomy, agribusiness and wildlife management, and a
fourth consultant brought in to provide technical support and assist in final document
preparation. Three USAID projects were evaluated concurrently by the evaluation
team, the CARE Livingstone Food Security Project (LFSP), the Zambia Wildlife
Authority's Administration Management Design (ADMADE) program, and the
Cooperative League of the USA (CLUSA) Rural Group Business Program. One
technical specialist was assigned to evaluate each project. In the CARE/LFSP case,
the agronomist was initially assigned full evaluation responsibility.
After a preliminary analysis of relevant documents along with general administration
activity in Lusaka, the agronomist and a local consultant traveled to Livingstone for
approximately two weeks to meet with project staff and beneficiaries. The first week
in Livingstone was spent meeting with project staff and visiting several Area
Management Committees (AMCs) in Kazungula and Kalomo districts. The AMCs
were selected to represent a range of levels of success in project implementation.
Date AMCs Visited
12-April Mukuni1
 and Jack3
13-April Delevu2
, Sekute1
, and Makunka3
14-April Sinde A3
, Katapazi3
15-April Mabwa4
, Mweemba3
16-April Shindu2
17-April Kasukwe4
Notes: 1 - weak AMC, 2 - average AMC,
3 - good AMC, 4 - new area.
The evaluation team also met with government officials while in Livingstone,
including the District Agricultural Officer, Ward Councilors, District Development
Coordinating Committee members, and the Permanent Secretary of Southern
LFSP Mid-Term Evaluation ARD
2
Province. During the later half of the second week, the evaluation team sponsored a
two-day workshop to solicit stakeholder views of LFSP. Participants at the workshop
included farmers, MAFF District personnel, community development personnel,
forestry personnel, representatives from the Southern Province Planning Section, and
CARE/LFSP staff. (For a complete list of people contacted for the evaluation, see
Annex I.)
A rough draft of the LFSP project evaluation was presented in late April to USAID
and project stakeholders. Following this presentation, it was agreed that the evaluation
did not adequately address the scope of work, and that additional fieldwork and
writing was needed. Subsequently a second evaluation team, consisting of an
environmental scientist and local agronomist, assumed responsibility for completing
the evaluation over an additional three-week period.
The second evaluation team first spent five days in Lusaka reviewing documents and
holding meetings with staff from USAID and CARE/Zambia. The team then traveled
to Livingstone for an additional five days, meeting with project staff and collecting
additional data. This team spent all its time in Livingstone and did not make any
additional community visits. One of the activities of this team was assisting the LFSP
Monitoring and Evaluation section to set up a database to process and analyze data
from the Community Self-Monitoring ledgers. Approximately 235 household records
were entered (see 6.1.4 on page 35).
A second draft of the evaluation report was submitted in late May, followed by a
second presentation for USAID and CARE stakeholders. Feedback from this
presentation was taken into consideration and the final report submitted to USAID on
5 June 2000.
3.0 BACKGROUND
3.1 Prelude to a food security crisis
During the first half of the 1990's, Zambia’s Southern Province experienced
several of the worst years of drought since independence, resulting in widespread
food shortages and relief efforts. Three long-term changes compounded the effects
of these droughts, setting the stage for the food crises:
ƒ removal of state marketing and input subsidies, leading to the collapse of
an agricultural economy based on hybrid maize
ƒ deterioration of the means of food production for small scale farmers,
including depleted seed stocks, and production assets such ploughs and
oxen
ƒ a long term trend in reduced rainfall in the region leading to a lower water
table and depleted supplies of surface water .
The interaction of these factors left the 1,000,000 people of Southern Province
highly vulnerable to food shortages when drought struck in the early 1990s.
LFSP Mid-Term Evaluation ARD
3
3.2 The Physical Environment
The three districts where LSFP operates, Kalomo, Kazungula, and Livingstone,
together comprise about 32,000 km2
. In 1996, Kalomo District was divided into
Kalomo District to the North and Kazungula District, to the South, with the
greater Livingstone area remaining a separate district.
Figure 1. Kazungula, Livingstone, and Kalomo Districts in Southern Province, Zambia
Source: USGS.
Topographically the area can be divided into a broad plateau area and valleys
along the major river corridors (see Figure 2).



Zimba
Kalomo
LIVINGSTONE
Elevation in m
600 - 705
706 - 839
840 - 948
949 - 997
998 - 1043
1044 - 1092
1093 - 1149
1150 - 1205
1206 - 1277
1278 - 1398
Rivers
0 20 40 60 80 Kilometers
Figure 2. Digital elevation model of Kalomo, Kazungula, and Livingstone Districts.
Source: USGS
Ecologically, the project area cuts across three zones based agroecological zones
defined by soils, topography, and rainfall:
LFSP Mid-Term Evaluation ARD
4
ƒ Kalahari – Sandy soils in the southern portion of Kazungula District, with
short grass and scattered mopane woodland. Rainfall represented by the
Livingstone Meteorological Station. Range during the last ten years from
400-750 mm/annum.
ƒ Plateau – Loamy soils with approximately equal calibration between sand,
silt and clay. Mainly found in Kalomo District. Flat, higher altitude (1,300
msl) than the Kalahari zone. Rainfall range during the last ten years
estimated between 650-900 mm/annum.
ƒ Escarpment – A smaller area mostly in Livingstone District with better
water supply due to higher rainfall and proximity to rivers. This is also a
hilly area with slopes of 10-60% inclination common.
Many of the technologies promoted by LFSP are targeted to specific zones based
on soil, topographic, and climatic requirements. Impact monitoring results are also
stratified by agroecological zone, as each of the AMCs have been placed into one
of the categories (see Figure 3).
Kazungula
Kalomo
Livingstone
Plateau
Kalahari Sands
Escarpment
Figure 3. LFSP Area Management Committees grouped by agroecological zone.
Source: LFSP.
3.3 Rainfall
Rainfall in Southern Province has probably always been erratic, but there has been
a noticeable long-term downward trend in total rainfall in the last 100 years. This
pattern was identified by farmers during a series of PRA exercises held in Kalomo
South in 1995, who perceived the rainfall decline accelerated in the 1970s and 80s
LFSP Mid-Term Evaluation ARD
5
(CARE, 1996). The long-term decline has resulted in wells and surface water
drying up more rapidly after the rains have ended. The decline in rainfall has been
noted in other parts of Zambia as well as Malawi and Zimbabwe, and is verified
by meteorological data (see Figure 4). Possible causes for the change in rainfall
include a broader regional climatic cycle or effects from deforestation.
Livingstone Metereological Station
Long Term Annual Rainfall
Linear regression, R2
 = 0.011
0
200
400
600
800
1,000
1,200
1,400
1,600
1904 1909 1914 1919 1924 1929 1934 1939 1944 1949 1954 1959 1964 1969 1974
mm
Figure 4. 70 year annual rainfall from Livingstone Meteorological Station.
Source: Nicholson, 1995
3.4 The People
Like much of Zambia, the LFSP area is sparsely populated, with approximately
8.6 people/km2
. In 1995, the rural population of Kalomo District was estimated to
be 205,000 inhabitants. It is estimated that the rural population in Southern
Province is increasing at a rate of 5% per annum.
The people in the LFSP area are predominantly Tonga speaking. In the southern
part of the LFSP area are predominately Toka-Leya, who belong to the Bantu￾Botatwe group. In the Kalahari sands, there is a large presence of people of Lozi
origin, who came from Western Province. Kalomo West is dominated by the
Plateau Tonga with some migrants from the middle Zambezi basin, mainly from
Sinazongwe area (Milimo, Kasonde-Ng'andu, Pongolani, Zulu, & Mwanza, 1997).
3.5 Land Tenure
The largest portion of land in the project area is held under the traditional land
tenure system, with use and occupancy rights allocated by chiefs. The GRZ
Commissioner of Lands cannot allocate any land under customary tenure without
first consulting the chief and Local Authority. To convert customary land to
LFSP Mid-Term Evaluation ARD
6
formal leasehold tenure, it is necessary to obtain consent from the chief and it is
also subject to the approval of the Commissioner of Lands.” According to DACO
of Kalomo, “90% of the land in the District is still governed under traditional
authority.”
3.6 Agriculture
The two districts where LFSP operates are characterized by low population
density, with land for farming purposes generally available for all. Livestock,
especially cattle, is a major component of the traditional farming system. The
Tonga ethnic group will leave the village and move to ‘greener pastures’ when
serious water shortages provide inhospitable conditions for people and livestock.
Figure 5. Demonstration of ripping using draught power in Siamasimbi AMC.
In Tonga-speaking culture, cattle represent both a production asset as well as an indicator of social
status. Source: LFSP.
The cropping system practiced is extensive agriculture with main cereal crops
being maize, sorghum and pearl millet. Total annual rainfall in some parts of the
project area can go below 600mm, conditions generally comparable with the
Sahelian zone of West Africa. In these areas, pearl millet and sorghum have a
comparative advantage over crops such as maize. Agriculture by small-scale
farmers in Southern Province is mostly rainfed, with irrigation being used for
vegetable, fruit and other specialty crops, only on small areas and primarily during
the dry season. The major food crops are produced under rainfed conditions.
Crops grown under rainfed conditions have critical periods when minimum
moisture conditions are absolutely essential, such as field preparation and
LFSP Mid-Term Evaluation ARD
7
planting, and crop development periods such as germination/emergence, stalk￾elongation, flowering and grain fill. Crop failure and low production can be
caused by both low overall rainfall amounts and poor distribution of rainfall
during the rainy season. Southern Province suffers from frequent micro-droughts
within a single season, see Figure 7 below.
Figure 6. Worksheet from Katapazi PRA, 1995.
Information from PRA exercises highlights farmer perceptions that the climate is becoming drier.
8
Rainy Season Rainfall 
Livingstone Meteorological Station
1995/96
0
20
40
60
80
100
120
140
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
10 day interval
Rainfall in mm
early 
Nov.
early 
May
total = 640 mm
Rainy Season Rainfall 
Livingstone Meteorological Station
1996/97
0
20
40
60
80
100
120
140
160
180
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
10 day interval
Rainfall in mm
early 
Nov.
early 
May
total = 737.1 mm
Rainy Season Rainfall 
Livingstone Meteorological Station
1997/98
0
20
40
60
80
100
120
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
10 day interval
Rainfall in mm
early 
Nov.
early 
May
total = 356.1 mm
Rainy Season Rainfall 
Livingstone Meteorological Station
1998/99
0
20
40
60
80
100
120
140
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
10 day interval
Rainfall in mm
early 
Nov.
early 
May
total = 592.7 mm
Figure 7. 1995-1999 Seasonal rainfall from Livingstone Meteorological Station.
Source: FEWS.
LFSP Mid-Term Evaluation ARD
9
3.7 LFSP Phase I, and the South-Western Drought Relief and
Mitigation Project
As a result of several consecutive years of poor rainfall, whose effects were
intensified by the economic and political changes discussed above, widespread
food relief was needed in Southern Province in 1991/92, 1994/95, and 1995/96.
CARE Canada began activities in Zambia in 1991, and served as the lead NGO
for food distribution in Kalomo south and Livingstone Districts.
The cost of mounting relief exercises year after year prompted CARE to take a
broader and longer-term view of food security amongst smallholder farmers. The
priorities for a new approach were ranked as:
ƒ develop an integrated program that promotes appropriate farming systems
ƒ improve methods of water harvesting and utilization
ƒ build the capacity of community institutions to enable them to plan
independently implement and manage activities
ƒ develop new private sector and market networks in an area previously
served exclusively by state or parastatal institutions
To begin exploring and addressing the underlying causes of food insecurity,
CARE implemented the Livingstone Food Security Project Phase I (LFSP-I) in
November 1994, with approximately two years’ funding from Canadian
International Development Agency (CIDA) and the UK Overseas Development
Agency (ODA). They continued to distribute food in the rural areas of the
Province even as the original LFSP project was being implemented.
In October 1995, CARE launched the South-Western Drought Relief and
Mitigation Projection (SWDRP) to manage the food relief distribution activity.
SWDRP was supported by USAID, ODA, CIDA, and UNICEF.
ODA - $325,000
Debt for Development - $176,000
USAID - $640,000
ODA - $675,000
CIDA - $440,900
UNICEF - K26,000,000
CIDA - $350,000
USAID - $3.6 million
DFID - $1 million (water)
2000
LFSP - Phase II
LFSP - Phase I
South-Western Drought Relief Programme
1994 1995 1996 1997 2001 1998 1999
Figure 8. Timeline of donor support for CARE's food security interventions in Southern Province.
Source: USAID Zambia.
LFSP Mid-Term Evaluation ARD
10
As can be seen in Figure 8, USAID's $3.6 million investment in LFSP Phase II
builds off of a substantial amount of non-US donor investment, totaling around $2
million USD. This amount was largely spent designing the program and
developing the project capacity to implement LFSP Phase II.
3.8 Livingstone Food Security Project Phase II
Building upon the lessons from LFSP-I, CARE decided to extend the project into
a second phase. A series of PRA exercises were conducted in March to May 1995
in Kalomo South to further probe the underlying causes of food insecurity in the
region and begin to develop intervention strategies. The PRAs identified the
following factors as the most important threats to food security:
ƒ lack of seed
ƒ lack of water for (humans and livestock)
ƒ lack of organization at the community level
ƒ depletion of natural resources
Figure 9. Women engaged in PRA exercise
LFSP was designed based on the results of a series of PRA exercises in 1995. Source: Robby
Mwiinga.
Based upon the results of the PRAs, CARE developed and submitted an
unsolicited proposal to USAID/Zambia in 1996 for LFSP-II. The operational
framework of LFSP is founded on three basic paradigm shifts from previous food
security interventions:
ƒ a shift from a focus on regional and national food security to household
and individual nutritional status
ƒ shift from food first (and only) perspective to livelihood perspective. The
livelihood approach focuses not only on the production of food, but also
the ability of households and individuals to procure the additional food
they require for an adequate diet.
LFSP Mid-Term Evaluation ARD
11
ƒ shift from materialist perspective focused on food production to a social
perspective which focuses on the enhancement of people’s capability to
secure their own livelihoods. (CARE, 1996)
The overall goal of the LFSP is to reduce vulnerability to drought
and achieve household food security by addressing the underlying
causes of vulnerability and food insecurity.
- LFSP Cooperative Agreement, 1996
4.0 PROJECT IMPLEMENTATION
LFSP is based on a livelihoods approach which focuses on strengthening the ability of
households to produce food and withstand drought. This holistic approach requires a
multi-faceted intervention program, focused around four operational objectives. These
four objectives serve as the nucleus of the primary lines of action.
Objective 1: Community institutions (CBOs) and capacity-building
Objective # 2:
Appropriate and
sustainable
farming systems
Objective # 3:
Improved water
harvesting and
natural resource
management
practices
Objective # 4:
Increased
incomes and
income earning
opportunities
Project Goal: Household food security and reduced vulnerability to drought
4.1 Project Components
4.1.1 Community Based Organizations
The establishment of community based organizations in LSFP defines the
operational context for all the remaining interventions. Capacity building of
CBOs is one of the characteristics that distinguishes a development project
from a relief effort. LFSP recognizes the importance of CBOs in implementing
activities and ensuring sustainability of the technologies and systems
introduced by the project. Indeed CBOs development is included as the first of
the four operational objectives, a status which elevates capacity building as a
end in itself as well as a means for achieving other objectives.
LFSP has introduced a multi-tiered CBO structure in the project area (see
Figure 11). Individual household members first belong to a farmer group, also
called a cell group, whose main function is managing the seed scheme.
LFSP Mid-Term Evaluation ARD
12
Representatives from these farmer groups in turn comprise a Village
Management Committee (VMC), whose main functions include managing the
seed scheme, maintaining the community self-monitoring ledger, and
coordinating extension training. Three to ten VMCs federate to form an AMC,
which coordinates the seed scheme and is the main entry point for project
extension services. The geographic size of AMCs approximates the size of an
administrative and political Ward.
The LFSP CBOs are integral elements of the extension and monitoring
systems, and greatly increase the effectiveness of LFSP field officers allowing
one field officer to work with approximately 1,000 farmers. They also serve as
the vehicle through which rural farmers can voice their opinions in the design
and management of project activities, and secure access to services. Presently
there are 54 AMCs and 598 VMCs in the three districts. Figure 12 below
shows the locations of the AMCs.
A newer addition to the LFSP CBO structure is the Interest Group (IG).
Interest Groups are focused around a specific income generating activity
(IGA), such as poultry or basket making, or a capacity building theme.
Members of an Interest Group fall under an AMC, however may not
necessarily be from the same VMC. Project guidelines limit the number of
Interest Groups that an individual farmer can join to three. Interest Groups
serve as a mechanism through which the project tries to build market linkages
and provide training. As the seed loan scheme becomes more established and
self-sustaining, many LFSP staff predict that farmer groups will gradually
evolve into Interest Groups, and Interest Groups will become the main focal
point of project activities.
Figure 10. Interest Group receiving training on mungongo nut oil extraction
The use of CBOs extends the reach of LFSP extension staff. Source: LFSP.
LFSP Mid-Term Evaluation ARD
13
Farmer/Cell Groups
• 4-7 members
• 1 group leader
• main activity is seed
scheme
Farmer Group
Village
Management
Committee
VMC
FG FG FG
Village Management Committees
• represent 8-15 farmer groups
• Chairman, Secretary, Treasurer,
Storekeeper
FG
VMC VMC
Area
Management
Committee
Facilitator
• elected by VMCs
• link for extension services
Facilitator
Area Management Committee
• represent 3-10 VMCs
• Elected Chairman, Secretary,
Treasurer, Storekeeper
Rural Households
Interest
Group
Interest Groups
• focused on a specific income
generating activity
• link for group marketing, training
Interest
Group
Interest
Group
Figure 11. LFSP CBO Structures
LFSP Mid-Term Evaluation ARD
14
6
6
6
6 6
6
6
6
6
6

 
 
 






 



 










  

 





 



 



 



 





7A
Mabwa
Kauwe
Nguba
Muzya
Songwe
Delevu
Kabuyu
Kasiya
Kayube
Libala
Maibwe Shindu
Katete
Luyaba
Dimbwe
Ng'andu
Mambova Mapanda
Sinde A
Sinde B
Milangu
Simango
Mulindi
Malimba
Busanga
Mweemba
Bbilili
Nantale
Kasukwe
Namukaba
Kantumbi
Ngwezi A
Ngwezi B
Sihumbwa
Muzumbwe
Simukombo
Sikaunzwe
Mweemba A
Nabusanga
Chabalanda
Mandandi B
Mandandi A
Siamasimbi
Siamundele
Dundumwezi
Mandia East
Mandia West
Kooma/mooka
Siakasipa 1
Siakasipa 2
Siandazya A
Siandazya B
Sipatunyana
Manyemunyemu
Musokotwane 1
Nyawa Central
Sekute Central
Mankunka Sekute
Kanchele Kalomo
Kanchele Kazungula
Makunka Musokotwane
Munyeke
Musokotwane 2
Mukuni Central
Kalomo
Zimba
LIVINGSTONE
District boundary
Rail Road
Roads
6 Health Centres
Rivers
 AMCs
10 0 10 20 30 40 Kilometers
N
W E
S
Livin
gstone Food Securit
y Project Area 
(AMCs
)
Figure 12. LFSP Area Management Committee locations
LFSP Mid-Term Evaluation ARD
15
4.1.2 Seed Scheme
Loss of local seed stocks, over-dependency on inappropriate hybrid maize
varieties, and the collapse of government subsidized input programs were
among the factors identified during the formative PRA exercises in 1995 as
negatively impacting food security. Subsequently LFSP's major intervention in
the project area has been the provision of early maturing and drought-tolerant
crop varieties that were previously unavailable to farmers or depleted after
years of repeated drought. The seed scheme was the focus of the LFSP's initial
entry into communities, and remains the backbone of the farming systems
section. The seed scheme improves household food security by serving two
different needs:
1. introducing drought resistant seed varieties
2. establishing local self-sustaining seed banks where farmers can access
seed supply on a loan basis
The seed scheme intervention both builds upon the work of community
capacity building, and also provides an activity focus for the establishment of
CBOs. The level of participation of communities in the management of CBOs
has been strengthened to the point that several AMCs are now managing their
own local seed banks with little or no assistance from the project.
The seed scheme operates by making seed loans to farmers. LFSP supports the
seed scheme by providing training on managing the seed scheme, sourcing
appropriate crops and varieties, and providing the initial injection of new seed
stocks. The logistics of seed distribution and repayment are a bit complicated,
but are captured in Figure 13 and Figure 14 below. The guiding principles of
the seed scheme are easier to follow and summarized in the box below.
Guiding Principles of the LFSP Seed Scheme
ƒ The scheme shall operate through groups only.
ƒ The members shall form groups on the basis of need and ability
to work together.
ƒ Groups shall comprise of approximately four to seven
individuals.
ƒ Village committees shall manage the groups and resolve
conflicts.
ƒ Village committees (VMCs) shall be managed by Area
management committees.
ƒ Groups will have to repay seed based on agreed rates to their
Village management committees, who will later take it to the
Area management committees.
ƒ Seed given is seed for multiplication.
ƒ The group leader/contact person is responsible to distribute
seed within his/her group.
ƒ Group to monitor the performance of the seed and crop
management practices.
LFSP Mid-Term Evaluation ARD
16
ƒ No one member should get more than seed of one crop seed at
a time.
ƒ In the event of failure by a member to repay, other members
will pay on behalf of the defaulter.
ƒ In the event of a specific crop not performing well, repayment
of seed crop can be made via another crop within the seed
scheme
ƒ Failure by group to pay hinders their future entitlements.
ƒ Groups will participate in agricultural experiments and other
activities as suggested by the Project (CARE).
ƒ Village Committees and groups to nominate individuals for
training.
ƒ Groups to establish what crop seeds they need for their group
members.
ƒ Group members should always be in contact with each other
through meetings.
The repayment rates for the seed scheme are summarized below. Additional
information about the seed scheme can be found in this report in sections 6.2
on page 45 and 10.2 on page 87.
Seed Loan Repayment Rates
1998/99 to present
Crop Loan
Amount
Expected
Repayment
Maize 10 kg 25 kg
Sorghum 4 kg 10 kg
Pearl Millet 2 kg 5 kg
Cowpeas 2 kg 5 kg
Ground nuts 10 kg 12 kg
Bambara nuts 10 kg 12 kg
Green gram 1 kg 3 kg
LFSP Mid-Term Evaluation ARD
17
Seed Delivery
Livingstone Food Security Project Seed Loan Scheme: Distribution
Farmer Group Leader
Request for seed
Household
Level
Farmer Group Meeting
• land and labor capacity
• last year supplies
• repayment record
VMC Secretary/
Chairperson
Written request for seed
Farmer Groups
Level
Village Management
Committee (VMC)
Level
Other Farmer Groups
VMC Meeting on Seed Requests
• Farmer Group repayment histories
• Community Self-Monitoring records
• farming capacity
• available seed stock
AMC Secretary/
Chairperson
Written request for seed
Other VMCs
AMC Meeting on Seed Requests
• VMC repayment histories
• available seed stock
• varieties loaned previously
Area Management
Committee (AMC)
Level
Sufficient Seed
Stocks? Yes
No
Request to CARE
Sufficient Seed
Yes Stocks?
No
AMC Storekeeper
Seed Delivery
VMC
Storekeeper
AMC
Secretary Record
keeping
• VMC Secretary
• Farmer Group
Leader
Record
keeping
Seed Delivery
Livingstone Food
Security Project
Level
graphic by Andy Lyons
USAID/Zambia
September 1998
Figure 13. Seed distribution.
LFSP Mid-Term Evaluation ARD
18
Seed repayment
Livingstone Food Security Project Seed Loan Scheme: Repayment
Households
That Received
Seed Loans
Farmer Group
Village Management
Committee (VMC)
Area Management
Committee (AMC)
Farmer Group Leader
VMC Storekeeper
• VMC Secretary
• Farmer Group Leader
Record
keeping
AMC Storekeeper
Other VMCs
• AMC Secretary/
Chairman
Record
keeping
AMC Seedbank
Surplus?
Yes
No
AMC Meeting
Return to
CARE for
Redistribution
to other AMCs
(discontinued)
Save for
another year
Return to VMC
members
Sell surplus
Profits
Construction
(e.g., seed storage,
dip tank)
Materials for AMC
(e.g., ledger books)
Support business
venture
Repayment Interest
Farmers must return the
same amount of seed they
borrowed, plus enough for
one other farmer. Ex. if
10kg was borrowed, 25 kg
would be repaid
graphic by Andy Lyons
USAID/Zambia
September 1998
Figure 14. Seed scheme repayment
LFSP Mid-Term Evaluation ARD
19
4.1.3 Improved Farming Practices
In addition to providing seed for improved crop varieties, LFSP has worked to
increase smallholder production with improved farming practices. These
efforts have focused on methods to improve soil fertility, retain soil moisture,
and prevent erosion. Practices for improving both field crops and vegetable
and fruit production have been developed.
The strategy for promoting improved farming practices is twofold. First
appropriate technologies which have been proven elsewhere are identified.
Secondly, the technologies are tested under field conditions and, based on the
results, either adopted, adapted, or dropped. The testing phase is conducted
through demonstrations and training of individual farmers. Subsequently,
promising technologies are supposed to be replicated on a wider scale, again
through training. Examples of technologies which have been promoted for soil
fertility improvement include green manure, crop rotations, cultivators, ridgers
and rippers (see Figure 5 on page 6). See Annex V for a complete list of the
specific improved farming practices promoted by the project.
Figure 15. Demonstration of intercropping maize with squash to conservation soil moisture.
Source: Robby Mwiinga.
4.1.4 Water Harvesting
As noted earlier, the area around Livingstone is experiencing a long-term
climatic shift, with average rainfall gradually falling and becoming
increasingly erratic. Droughts and dry spells within the rainy season are
increasingly common, and surface and underground water supplies are much
reduced and more ephemeral compared with 30 years ago. Lack of surface
water was identified during the series of PRA exercises during LFSP-I as one
of the major threats to food security in the region.
LFSP Mid-Term Evaluation ARD
20
LFSP and the local people have tackled the issue of improved water
availability and utilization in various ways. This has included construction and
rehabilitation of dams, wells and boreholes. Village water, sanitation and
hygiene committees (V-WASHE) have been established in association with
VMCs and have promoted the healthy management of domestic water points.
The benefits of these initiatives include empowerment of women, who tend to
take a strong role in water management institutions.
4.1.5 Natural Resource Management
The application of NRM methods in LFSP was initially focused on the control
of erosion near water points to reduce siltation and improve water quality.
People understand and appreciate the fact that diminishing water resources are
linked to catchment degradation. Through training and demonstrations, LFSP
has helped community water committees reduce the amount of sediment
flowing into water retention areas, and effectively increase the period of
usability for seasonal water points.
NRM technology is now being expanded to broader watershed management
and the sustainable harvesting of forest products. The strategy used is again a
combination of training, small-scale pilot testing, and then replication. Taking
advantage of the provisions of the new Forest Act, which provides for joint
forestry management in Zambia, LFSP has also started to create forest
management awareness and institutional capacity development in
collaboration with other stakeholders.
4.1.6 Small Economic Activity Development
The Small Economic Activity Development (SEAD) component of the project
is designed to improve food security by increasing the asset base of farmers to
withstand drought years, and provide an alternative means for obtaining food.
The SEAD component initially was focused around a household savings and
credit program, modeled after a similar scheme from an urban development
program. However the savings program stagnated after the initial initiation
period. Experiments with micro-credit for vegetable production were positive
but were on a pilot basis only and not repeated.
Today the project is focusing on developing linkages with small-scale group
enterprises and urban markets. This appears to be a much more demand driven
service, and may provide much needed infusions of cash into the LFSP
communities. The project is experimenting with a host of market linkages,
including vegetable production, thatching grass, and milk. The project is not
looking for a universal enterprise model that will work throughout the entire
project area, but to use a strategic advantage analysis of the most appropriate
market linkage for each community based on agroecological variables and
access to urban markets.
Some of the more successful pilots include an outgrower scheme based on
maize, dry season vegetable gardening, and curios production. Other
LFSP Mid-Term Evaluation ARD
21
promising linkages currently being explored include dairy production and non￾timber forest products such as thatching grass. To support the development of
market linkages, the project is also supporting its CBOs with business training
and market research.
Figure 16. Vegetable garden at Mandia
Providing loans for vegetable production in Mandia was successful in raising rural income but
can't be replicated everywhere.
4.2 Project Resources
4.2.1 Staffing
To implement the diverse range of activities required by the improved
livelihoods strategy, LSFP is focused around four programmatic sections, each
led by a coordinator. Backing up these sections are the Project Manager and
Assistant Project Manager, and administrative and technical units (see Figure
17). Currently the SEAD Coordinator and Marketing Coordinator positions are
vacant. The total number of project staff when all positions are filled is 50,
including nine staff in Kalomo funded by DFID.
The project headquarters is based in Livingstone, with a separate field office
in Kalomo. Sixteen staff are based at the Kalomo office, while the remaining
operate out of Livingstone. All extension officers reside in either Livingstone
or Kalomo, but have motorcycles to travel to the AMCs.
22
Assistant Project
Manager
Project Manager
Farming
Systems
Coordinator
Extension
Officers (6)
Livestock
Coordinator
SEAD /
Marketing
Coordinator
Technical
Coordinators
(2)
Water
Technicians
(4)
SEAD
Officers (2)
Monitoring &
Eval
Coordinator
Senior
Accountant Documentalist Senior Program Administrator
Senior
Admin.
Assistant
Procurement
Officer
Secretaries
(2)
Stores
Officer
Mechanics
(2)
Drivers
(4)
Office Help
(4) Receptionist
Training
Officers (2)
Gender
Analyst
Monitoring
& Eval
Assistant
Accountant
Accounts
Assistant
Figure 17. Organizational Chart of LFSP.
Source: Administration Coordinator, LFSP.
LFSP Mid-Term Evaluation ARD
23
4.2.2 Vehicles
At the beginning of USAID financing, the project inherited two older 4-wheel
drive vehicles from on–going Southwest Drought Relief Project. The project
also was given an older (1992) 4-wheel drive vehicle from CARE Office in
Lusaka. In November of 1996, a month into the first activities under USAID
funding, the project procured two new 4-wheel drive Toyota Land cruisers
with USAID funds. With use of the older vehicles, the project was able to wait
until November/December 1998, before procuring the last two vehicles with
USAID money. The project bought one 4-wheel drive Toyota pickup and one
4-wheel drive Toyota Land cruiser with these funds, thus closing out the
vehicle line item in the project budget. In 2000, a ten-ton truck was purchased
under an approved budget amendment for unused funds.
4WD from SWDRP - sold
4WD from SWDRP - sold
4WD from CARE-Lusaka - sold
4WD landcruiser - USAID funds
4WD landcruiser - USAID funds
4WD landcruiser - USAID funds
4WD pickup - USAID funds
4WD pickup - DFID funds
Ten-ton truck - USAID funds
1996 1997 2001 1998 1999 2000
Figure 18. Timeline of vehicle procurement in LFSP.
Source: LFSP Administration.
Use of the vehicles seems well organized and restricted for moving staff where
they need to go for fieldwork. Rarely are vehicles used without full occupancy
by staff. The coordination of a diverse set of project activities using a
restricted number of vehicles, a large staff, and two offices reflects the strong
capacity of management and adherence to operational policies.
4.3 Financial
4.3.1 Spending by category
The LFSP's five-year budget from USAID totals $3,647,000 of which
$276,000, or approximately 7.58%, is CARE USA overhead. The largest
budget category is personnel, reflecting the large staff size and competitive
compensation levels. The second largest categories are vehicles (15%) and
equipment (20%).
LFSP Mid-Term Evaluation ARD
24
LFSP Total Expenditures of USAID Financing 10/96 - 09/01
Total = $3,631,403
Staff Training
2%
Consultancy
2%
Miscellaneous
1%
CARE USA Overhead
7%
Office Support6
8%
Equipment & 
Activities5
20%
Travel4
6%
Vehicles3
15%
Personnel Shared2
5%
Personnel Direct1
34%
Figure 19. Total expenditures (actual and projected) by category for LFSP 1996-2001.
Source: CARE Zambia.
Notes: 1 - Livingstone staff salaries
2 - shared direct costs, including portions of two expatriates salaries at head office
3 - vehicle purchases, fuel, and maintenance
4 - bus fares, airfares, per diem for staff travel
5 - computers, office equipment, seed, all materials for project activities including
training
6 - stationery, communications, rent, utilities, insurance
4.3.2 Spending by Year
Through 20 February 2000, LFSP spent $1,866,000. The average monthly
expenditure from July 1996 to June 1999 was $38,154. This average was
significantly lower than projected, caused in large part to the use of $350,000
from CIDA funds during the first year of the project. In 1999, the project
started spending more money under a plan approved by USAID/Zambia
(LFSP, 1999d). Between July 1999 to February 2000, the average monthly
expenditure increased to $61,541. Under the current burn rate, the project is
expected to exhaust the financing as scheduled by the Close of Project date of
September 2001.
LFSP Mid-Term Evaluation ARD
25
Livingstone Food Security Project
Total Project Funds Committed and Total Expenditures
$0
$200,000
$400,000
$600,000
$800,000
$1,000,000
$1,200,000
Yr. 1 Yr. 2 Yr. 3 Yr. 4* Yr. 5*
Projected Actual *estimated
Figure 20. LFSP total funds committed and expenditures by year.
Source: CARE Zambia
Currently the project is on target with the forecasted spending per budget
category. If expenditures continue as planned, it will wind up over spending
on personnel by about 13%, but compensate by under-spending slightly on
several other categories. See Figure 21 below.
Livingstone Food Security Project
Project Expenditures and Actuals By Budget Category
$0
$200,000
$400,000
$600,000
$800,000
$1,000,000
$1,200,000
$1,400,000
$1,600,000
Personnel
Vehicles
Travel
Training
Equipment
Office Support
Consultancy
Miscellaneous
Budgetted Actual
Figure 21. Total projected expenditures and actuals by budgeted category.
LFSP Mid-Term Evaluation ARD
26
4.3.3 Internal/External Funding
Within Zambia, approximately 19% of USAID financing is used to support the
headquarters office in Lusaka, including the now defunct MERU. Of the
remaining 81% spent by the Livingstone office, 61% is spent on field
operations and 20% is spent on office overhead. Thus a total of 39% of
USAID financing expended in Zambia for LFSP is spent on operating the two
offices in Lusaka and Livingstone. When combined with the 7.5% committed
to CARE/USA, about 43% of the total financing is spent on administration
and overhead. Although this percentage may seem high, what it buys is the
corporate capabilities a large international NGO like CARE can deliver,
including a reasonable level of confidence that project deliverables will be
met, the ability to leverage project lessons into policy change (e.g., see 9.4 on
page 81), external auditing, technical support, and the ability to solicit other
donor funds (see Figure 8 on page 9).
Livingstone Food Security Project
Within Zambia Expenditures of USAID Financing
Lusaka
19%
Field operations
61%
Office Overhead
20%
Livingstone
82%
Figure 22. LFSP expenditures within Zambia, 1996-2000.
Source: LFSP, Management Accounts.
4.3.4 Cost per beneficiary
The LFSP financial office can not easily estimate the cost of specific services,
because many project expenditures are used to support multiple services.
However the overall cost per beneficiary is decreasing as shown in Figure 23.
This is due primarily to a continuous increase in the number of beneficiaries
each year with roughly stable expenditures. Year 1 should not be considered in
the analysis because it included $350,000 of funding came from CIDA which
has not been incorporated in the cost calculation. The cost per beneficiary
remained roughly stable between year 3 and year 4 due to the graduation of
AMCs, which lowered the number of beneficiaries. However once the newly
formed AMCs are added to the equation, the cost per beneficiary should drop
further.
LFSP Mid-Term Evaluation ARD
27
Livingstone Food Security Project
Overall Cost Per Beneficiary
$0
$2
$4
$6
$8
$10
$12
$14
Yr. 1 Yr. 2 Yr. 3 Yr. 4
$/person
Figure 23. LFSP overall cost per beneficiary.
Cost per beneficiary = total annual expenditures from USAID on field operations / total
number of beneficiaries. Source: LFSP Accounts Management.
5.0 BENEFICIARIES
Beneficiary Profile
As described earlier, the primary beneficiaries of LFSP are smallholder farmers, who
happen to be mostly Tonga, in Kazungula, Kalomo, and Livingstone districts. As
illustrated in Figure 24 through Figure 28 below, the majority of participating
households have 10 members or less, and cultivate a total area of around 2 ha. There
is a positive correlation between family size and area cultivated, suggesting that this is
an extensive farming system where labor is one of the main limiting factor. There is
no major gender imbalance for any of the age classes, however unlike many
demography profiles in Zambia there is no increase in population in the upper age
classes. This suggests that there isn't a significant return of retirees from urban areas
back to the village, as is seen elsewhere in Zambia. The majority of households fall in
the lowest two wealth classes with a total asset base valued at $200 or less.
LFSP Mid-Term Evaluation ARD
28
CSM - Household Size Histogram
0
5
10
15
20
25
 1: 2 3: 4 5: 6 7: 8 9:10 13:14 15:16 23:24
Family size
Number households
Figure 24. Household size distribution of project beneficiaries, 1998.
Source: CSM database.
CSM - Average Area Cultivated by Household
0
0.5
1
1.5
2
2.5
3
3.5
4
escarpment kalahari plateau
Area cultivated ha
1996 1997 1998 1999 2000
Figure 25. Average area cultivated per household for all crops combined.
Source: CSM database.
CSM - Total Area Planted Vs. Family Size
R2
 = 0.2935
0
2
4
6
8
10
12
14
16
18
02468
# household members ages 10-60
ha
Figure 26. Relationship between family size and area under cultivation, 1998.
Source: CSM database.
LFSP Mid-Term Evaluation ARD
29
CSM - Household Age Structure
0
50
100
150
200
250
300
1:10
11:20
21:30
31:40:00
41:50:00
51: 60
61: 70
71: 80
91:100
female male
Figure 27. Age and gender profile of LFSP households, 1998.
Source: CSM database.
CSM - Value of All Household Assets
0
10
20
30
40
50
60
70
80
90
100
 1: 100
 201: 300
 401: 500
 601: 700
 801: 900
1001:1100
1201:1300
1401:1500
1701:1800
1901:2000
Approx value of assets in US$
Num households
Figure 28. Wealth profile of households in LFSP, 1998.
Source: CSM database.
Number of beneficiaries
The number of direct beneficiaries is calculated either by a physical count from the
membership lists of VMCs, seed distribution lists, or estimates based on the number
of VMCs. As can be seen in Figure 29, the number of direct beneficiaries has been
steadily increasing. LFSP has been able to reach such a large target population
because of the increased efficiency of working through CBOs. In addition the CBOs
that were formed by LFSP, each AMC has also formed an average two VMCs
independently of direct project intervention (LFSP, 1999a).
Indirect beneficiaries are considered to be all members of the household of farmers
participating in the seed scheme or one of the other project activities. The project
usually uses an estimate of 6 people per household to calculate indirect beneficiaries.
Other multiplier effects of LSFP interventions, such as distribution of the improved
seed varieties to non-participants, have been noted by project staff but difficult to
LFSP Mid-Term Evaluation ARD
30
measure. Milimo (1999) found that seed diffusion can extend up to 50 km from LFSP
VMCs (see Table 9 on page 91), however he didn't try to estimate the number of these
non-participating beneficiaries.
Livingstone Food Security Project
Number of Direct Beneficiaries
0
5,000
10,000
15,000
20,000
25,000
30,000
1994/95 1995/96 1996/97 1997/98 1998/99 1999/00
Figure 29. Number of direct beneficiaries by year
Source: LFSP. Direct beneficiaries are counted by VMC membership records, or if those are not
available then seed scheme records.
Recently, the project established seven new AMCs. The purpose of this expansion
within the existing district was to fill in the gaps in the service area and apply the
lessons learned from the previous efforts in CBO capacity building. Figure 30 below
shows the distribution of the new AMCs, which for the most part are at the edges of
the project's service area or "in-fill" areas.
LFSP Mid-Term Evaluation ARD
31
Location of LFSP AMCs Established in 2000
= new AMC
Figure 30. Location of new AMCs.
Figure 31. Dundumwezi PRA, 1998.
Entry into a new area begins with a PRA to collect information about the area, orient people to the
LFSP CBO structure, and begin the capacity building process.
LFSP Mid-Term Evaluation ARD
32
6.0 PERFORMANCE INDICATORS
The log framework in the LFSP cooperative agreement between CARE International
and USAID/Zambia lists 32 anticipated results which should be monitored. These
performance-based results are grouped into four categories, corresponding to four
program objectives. There is also an indicator for the overall objective of improving
food security and reducing vulnerability to drought.
LFSP reports on its performance indicators in its quarterly and annual reports. This
section presents a summary of each of the performance indicators as reported by these
quarterly reports, supplemented in some cases with additional data.
6.1 Community institution and capacity building
6.1.1 Within 1 year, 9,000 farmers organized in seed groups and VMCs
will be receiving, multiplying and repaying seed loans according to the
established seed scheme.
RESULT
ACHIEVED 100% 
The results for this indicator is discussed together with 6.1.2 below.
6.1.2 Within 3 years, 18,000 farmers organized into seed groups. VMCs
and area communities will have participated in the seed scheme.
RESULT
ACHIEVED 100%
Figure 32 below shows the number of VMCs and AMCs over the course of the
project. These CBO structures were developed to serve multiple purposes, but
were designed for management of the seed multiplication scheme. This has
allowed the project to extend its reach in improving seed supply for farmers
and reach its targets for the number of farmers in seed groups as shown in
Figure 29 on page 30.
LFSP Mid-Term Evaluation ARD
33
Number of AMCs
0
10
20
30
40
50
60
1996/97 1997/98 1998/99
Figure 32. LFSP AMCs per year.
Source: LFSP.
Table 1 below shows the repayment rates for the seed scheme for 1996/97.
Repayment is presented both as a percentage of tonnage of seed returned, as
well as a percentage of the total value of the seed. These two methods of
calculating repayment rates don’t produce different results for individual
crops, but has an effect in determining the overall repayment rate. The
difference between the two methods in the overall repayment rate is due to the
fact that groundnuts have the highest seed value at $2,300/Mt, and the
groundnut repayment rates was extremely low for 1996/7 due to poor yields
especially for Natal Common.
Crop Expected Repayment Rate
repayment
Mt
Actual
repayment
Mt
Value
$/Mt percent
tonnage
percent
cash value
maize 25.8 23.4 1,090 90.5 90.5
sorghum 32.4 32.6 1,015 100.6† 100.6†
millet 4.3 2.7 901 62.6 62.6
cowpeas 3.6 1.8 1,520 50.8 50.8
groundnuts 5.8 0.3 2,300 4.5 4.5
Total: 71.9 60.7 84.4 61.4
Table 1 Repayment rates on seed loans to VMCs & AMCs in 1996/7.
Source: LFSP
† percentage returned for some crops may be greater than 100 because farmers were allowed to
substitute repayment seed from a poorly yielding crop with the surplus seed from another crop
6.1.3 Within one year, an overall community institutional structure will
have been agreed upon with communities and accepted by GRZ extension
organizations.
 RESULTS
COMMUNITY ACCEPTANCE OF CBO STRUCTURES: YES
 GRZ EXTENSION ACCEPTANCE: PARTIAL
LFSP Mid-Term Evaluation ARD
34
The structure and function of CBOs in the LFSP has been the focus of several
evaluations (Hatwiinda & Whitehead, 2000; Milimo et al., 1997; Milimo &
Tripp, 1999; Ndiyoi, Bwembya, & Kasuta, 1999; Turner, 2000). These
reviews note that there is some variation in the structure and functioning of
CBOs, which depends on age and leadership quality. However many AMCs
have developed a degree of autonomy, authority, and initiative, which are
important qualities for sustainability and criteria for graduation.
That the communities served by the LFSP should accept the structure of the
CBO model is no great surprise. The CBOs are also the platform through
which community residents have input into project and receive valued
services, namely access to improved seed varieties, training in farming
systems, and improved water harvesting structures. The three-tiered model of
village cells/VMCs/AMCs (see Figure 11 on page 13) was developed in close
consultation with the local communities, if not initiated by them. LFSP also
invests heavily in providing training to leadership of the CBOs to ensure that
the roles and functions of the CBOs are understood and adhered to.
Government extension organizations have been slower to embrace the LFSP
model of CBOs. Officially MAFF supports working through community
groups at both the national and district levels. All stakeholders spoken to
stated that the MAFF policies are conducive for closer collaboration between
LFSP CBOs and MAFF extension officers. See 8.1 on page 74 for additional
details on the relationship between LFSP and MAFF.
However most people will also admit that the implementation of this policy
into practice at the district level has generally not gone as smoothly as was
expected. There are two schools of thought on the causing for the slow
progress. Representing the perspective of MAFF, extension officers state they
were not very integrated into the planning and capacity building of CARE
CBOs, particularly during the early phases of the program. Although MAFF
was represented during the three 1996 PRA exercises which were conducted
to design the strategies of LFSP, their participation in the implementation of
the program was limited.
MAFF also already had farmer groups formed in the LFSP area that their
extension officers worked with through the T&V extension model (see section
10.1 on page 84). Although these groups did not have the capacity building
training that LFSP AMCs are taught, MAFF officers state they did not
understand why CARE choose to introduce a completely new set of CBO
structures instead of building on the existing groups. MAFF officers add that
their interactions with LFSP AMCs were also constrained by resources,
particularly transport. Some MAFF officers might also state that CARE
presented a threat to their operations, stealing both target communities as well
as field staff. Therefore from their point of view, a strategy of demarcating
separate areas of operation seemed to be the best approach.
LFSP Mid-Term Evaluation ARD
35
LFSP staff have a different perspective to the question of MAFF involvement
with their CBOs. From their point of view, MAFF officers may have found it
difficult to embrace the CBO approach because they did not feel comfortable
with or convinced of the value of participatory extension methods. They note
that extension training in MAFF has a long history of being supply-driven and
focused on interactions with individual farmers. MAFF officers may also have
been unmotivated due to their low pay scale and unwilling to work with LFSP
staff without lunch allowance.
Undoubtedly all of the above factors identified in both schools of thought
contributed to the low level of participation of MAFF extension officers in
LFSP. The institutional relationship was also most certainly affected by
personality dynamics. All would concede that the relationship was not helped
by the lack of a formal mechanism such as an MOU defining the operational
relationship between the two institutions. In fact an MOU was drafted, and
might have even been signed at the national level, but according to CARE
staff it was general in nature and never used to guide the relationship between
the two institutions at the district level.
However this is not meant to suggest that there was no cooperation or
interaction between the two organizations. Indeed there have been several
MAFF extension officers who have been quite active with LFSP activities,
including trainings and water projects. However it would be accurate to
characterize these cases as fairly limited and based on individual interests as
opposed to a strong institutional relationship.
Recent developments suggest a brighter future for the relationship between
MAFF and LFSP may lie ahead. Both organizations acknowledge it is in their
strategic interests to work closer together. LFSP is entering a phase where they
need to scale up proven methods for food security and prepare for a possible
phasing out of project services. MAFF officers recognize the advantages of
working with LFSP CBOs which have the capacity to spread extension
messages and test improved agricultural technologies. In 1999, MAFF and
LFSP coordinated plans to support the graduation of ten AMCs. The DACO
transferred five of his strongest field staff to the areas where AMCs were
graduating, and LFSP helped to train the officers in participatory extension
methods. Preliminary results from the graduation process suggest that future
graduating AMCs will be best served by an even greater period of overlap
between LFSP and MAFF field officers.
6.1.4 Monitoring and information systems established with community
institutions for tracking their performance.
 RESULTS
M&I SYSTEMS IN PLACE: YES
USE OF M&I SYSTEMS: ROOM FOR IMPROVEMENT
LFSP Mid-Term Evaluation ARD
36
Participatory Data Collection Methods
LFSP uses various methods to collect information about the performance and
impact of project interventions. PRA exercises and topical appraisals provide
in-depth information about livelihood strategies, community dynamics, and
farming systems in the project area. These types of exercises are initiated and
facilitated by LFSP staff, and used primarily for planning purposes (as
opposed to impact monitoring). The project also relies heavily on qualitative
and quantitative feedback from extension officers who are required to write
monthly activity reports as well as summaries of all community meetings and
trainings.
Food Production Trends Survey
Still at the project level, the LFSP Monitoring and Evaluation Unit conducts
an annual household food security survey called the Food Production Trends
Survey. This survey is administered to the same 220 households each year in
all three agroecological zones, who were randomly selected in 1996 to
represent the entire population of beneficiaries. This dataset is not used by
communities, but is used to study impact issues at the project level.
Community Self-Monitoring Books
To monitor project impact at the community level and provide information
tools for community level planning, the project introduced the Community
Self-Monitoring (CSM) ledger. These ledgers are maintained by the VMC
chairman or secretary, and record data on food security indicators from an
annual door to door survey of households. Information recorded includes
household size, household composition, assets, agricultural production, the
expected period of food availability, and anticipated strategies for coping with
food shortages (Lyons, 1998a). The CSM ledgers are the primary monitoring
and information tools at the community level.
While the CSM ledgers represent an innovative method for gathering
information at the community level, and the project has made notable progress
in training communities to use CSM ledgers, there is still room for
improvement in utilization. Neither the project nor the CBOs fully utilize the
information in the CSMs. A 1998 study tour found that many AMC and VMC
officers didn’t really know what to do with information in the CSM ledgers,
other than to use it to document their own poverty in the hopes of getting more
development assistance (Lyons, 1998a). However there are other examples of
using the CSM including the organization of food relief operations.
A recent study by the LFSP Monitoring and Evaluation section found that
slightly less than half of the AMCs use information from the CSM (see
Figure 33). "Use" was defined as the presence of one or more of the following
behaviors:
ƒ using CSM data to offer guidance on production strategies to farmers
at the time of enumeration
LFSP Mid-Term Evaluation ARD
37
ƒ using CSM data to manage the seed scheme, including collection of
seed loans
ƒ using CSM data to plan the distribution of food relief
Livingstone Food Security Project 
Percentage of AMCs Using Community Self-Monitoring Ledgers
0
10
20
30
40
50
60
1996/97 1997/98 1998/99
0%
10%
20%
30%
40%
50%
60%
70%
Total AMCs AMCs Using CSM % AMCs using CSM
Figure 33. Percentage of AMCs using Community Self-Monitoring Ledgers
Source: (Hatwiinda & Whitehead, 2000)
Although most CBO members will probably state it is important to keep
records, the CSM appears to still be primarily project driven. However
ironically, the project has also not made full use of the CSM. This is partly due
to the fact that the ledger books stay in the villages. This demonstrates the
importance of community ownership of data but also makes it difficult for the
books to be analyzed except on an ad-hoc basis.
The CSM is one of LFSP's more innovative tools for monitoring and
evaluation, and has attracted outside attention. In 1999, LFSP hosted a
regional workshop on participatory monitoring and evaluation on behalf of
PELUM, a regional organization. Twenty-six participants from four countries
participated.
CSM Database
For this evaluation, a new database was set up by the M&E unit to enter and
analyze data from the CSM ledgers. The evaluation team assisted the M&E
staff in designing the database which was then used to enter approximately
235 records for households from all three ecological zones (see
Figure 34). Analyses from this dataset are presented throughout this evaluation
report. If LFSP continues to enter data into the CSM database, it could make
an important contribution to future analyses of project impact.
LFSP Mid-Term Evaluation ARD
38
CSM - Number of Households with Production Data
0
50
100
150
200
250
escarpment kalahari plateau
1996 1997 1998 1999 2000
Figure 34. Amount of data currently entered into the Community Self-Monitoring database.
Source: CSM database.
As can be noted from Figure 34, with the exception of the plateau zone, the
current sample sizes in the CSM database are relatively small, particularly for
the earlier years. This is not a result of poor record keeping practices at the
VMCs, but the time constraints for data entry during the evaluation period.
But as a consequence, all multi-year trend CSM analyses presented in this
report must be interpreted carefully, and extrapolations to the general
population made with caution. On the other hand, analyses based on data from
a single year, such as the household demographic profile in Figure 27,
represent a fairly large sample sizes and are less likely to be affected by
sampling bias.
Seed Scheme Records
Separate from the CSM and Food Production Trends Survey, data is kept at
the AMC level on the seed loan and multiplication scheme. Record keeping
was taught during CBOs training exercises, and proper records is a
requirement to get new inputs of seed from LFSP. The information is used by
the AMCs to keep track of how much seed they have given out, and is also
used in the collection of seed loans and allocation of new loans. The project
uses the seed scheme records to estimate the number of direct beneficiaries,
get statistics on gender balance, and record amounts of seed being distributed
in the project.
Data Quality in LFSP Monitoring Systems
In any monitoring system there is of course a concern that data is accurate
(e.g., unbiased) and valid (provides appropriate information to answer the
question of interest). Data quality can be affected by numerous factors,
including sampling, intentional bias, and measurement consistency.
The quality of information that comes out of participatory data collection
methods, such as results from PRAs and topical appraisals, is generally
LFSP Mid-Term Evaluation ARD
39
trustworthy, but affected by the skill of the facilitators guiding the exercise. If
the focus of the exercise is too heavily focused around intervention planning,
for example, community residents will naturally try to provide information
which will maximize the likelihood they will receive some services. Likewise
if the agenda of the facilitator is clearly evident from the line of questions,
community residents will often try to tell them what they perceive is wanted.
LFSP tries to avoid such issues by dividing PRAs into a diagnostic phase and
negotiation phases. CARE also has a strong reputation for developing many of
the techniques used in participatory data collection, and the LFSP staff are
well versed in participatory techniques. The range of exercises conducted in
PRAs, such as food calendars (see Figure 35), social mapping, resource
mapping, and pair-wise ranking of food security threats, are general enough to
yield trustworthy results. The use of groups also tends to buffer out individual
voices that would purposefully misrepresent reality, however these same
forces can suppress the opinions of less powerful groups such as women. That
is not to say there isn't any bias introduced into PRAs by virtue of the fact that
LFSP has a specific set of services to offer, however the big findings such as
the underlying production patterns and threats to food security are almost
certainly valid.
Figure 35. Food calendar from Makunka PRA, 1995.
The Food Production Trends Survey is a valuable dataset on household level
indicators, but vulnerable to two types of data quality errors. First, because
this is a quantitative dataset, it is particularly susceptible to errors stemming
from sample size and sample selection. The respondents for the survey were
initially selected at random, from VMC membership lists. Therefore they
should comprise a representative sample of beneficiary households. Although
the total sample size of 220 households was selected to provide an adequate
sample size, subsequent surveys altered the sampling strategy when heads of
households were not at home or had moved. The bias introduced by this loss
LFSP Mid-Term Evaluation ARD
40
of respondents is difficult to measure, but it makes trend analyses less reliable.
See 10.3.4 on page 96 for a more thorough discussion of other sampling issues
in the Food Production Trends Survey.
A more serious problem with the Food Production Trends Survey is the
possible under-reporting of production because of the belief that the results
may be used to plan food aid. Questionnaires administered by outsiders from
development or relief organizations are most susceptible to this type of bias,
because they are not in the position to verify most responses and are perceived
to be possible sources of assistance. Production data also fails to capture the
amount of crop that was eaten from the field before harvesting, either as a
coping strategy against hunger or to sell for cash needs.
The Community Self-Monitoring books are less susceptible to deliberate
under-reporting because they are administered by a local person who has some
knowledge of the assets and production levels for local households. However
the CSM does have some problems with consistency and standardization
across the project. For example, in the household assets section, some VMC
chairmen record assets as small as chairs, buckets, beds, etc. However other
VMC chairmen only record large assets and production assets such as ploughs,
oxen, etc. This makes it difficult to compare total asset levels between
different areas.
As a general strategy, data quality issues can be mitigated by examining
multiple indicators from multiple data sources for the same impact question.
LFSP makes use of this triangulation strategy for planning systems, but needs
to do more of this for impact monitoring. For example the FPTS and CSM are
almost identical in scope, but have not been often analyzed together to look for
correlation between the indicators.
6.1.5 Within 3 years, members of all AMCs will have received
management training.
RESULTS
AMCS WHICH RECEIVED MANAGEMENT TRAINING: 100%
As illustrated in Figure 32 on page 33, the number of AMCs in the project
increased from 34 in 1996 to 56 in 1999. Part and parcel of the process of
establishing an AMC is management training in CBO leadership. The project
requires attendance at management training as a prerequisite for AMC
establishment and participation in project activities such as the seed scheme.
Hence all AMCs have received training at the time of their formation.
LFSP Mid-Term Evaluation ARD
41
6.1.6 Within 3 years, AMCs will be coordinating mult iple program
activities in their areas
RESULTS
AMCS COORDINATING MULTIPLE PROGRAMS: YES, BUT NOT ALL
A recent internal evaluation found that 64% of CBOs are planning,
implementing, and managing development activities by CARE (Hatwiinda &
Whitehead, 2000). This is higher than the internal target of 60%, but still far
less than the ideal goal of 100%. The remaining AMCs are either not very
active or remain dependent on LFSP staff to plan, implement, and manage
development activities.
6.1.7 Within 4 years, responsibility for all activities will be decentralized
from project staff to community institutions
RESULTS
DECENTRALIZING: SEED MULTIPLCATION BECOMING DECENTRALIZED
GRADUATING AMCS OFF TO A GOOD START
In 1999, the project developed a set of criteria to determine which AMCs were
ready for 'graduation' (see box below). Based on these criteria, 10 of the 44
AMCs were graduated. Graduated AMCs still receive occasional visits from
LFSP extension staff, but manage most of their activities, including the seed
scheme, on their own.
Five of the ten AMCs were linked up with MAFF extension officers at the
time of graduation. Preliminary results reveal that the AMCs which were
linked up with MAFF agents show less initiative and are less active than the
AMCs which were left on their own. Anecdotal evidence suggests that this
lack of initiative may result from dependency on the MAFF officer for
guidance and facilitation. However the pattern in not universal, and there are
some MAFF officers who seemed to have truly internalized the participatory
facilitation approach and act more as enablers as opposed to external service
providers.
Criteria for AMC Graduation
ƒ Written constitution understood by at least 60% of
participants
ƒ Meetings held at least once a month
ƒ Meeting agenda to include non-LFSP issues
ƒ Meeting attendance at least 60%
ƒ Minutes kept for all meetings
ƒ Additional training received after core CBO training
ƒ All topics from Training of Trainers course covered
ƒ At least 60% of CBO participants attended training
ƒ AMC visitations to VMCs at least once a month
ƒ Non-LFSP issues discussed on VMC visits
LFSP Mid-Term Evaluation ARD
42
ƒ Accurate record keeping for all activities
ƒ At least 10% of CBO members joined since initial
formation
ƒ Demonstrated examples of resource mobilization
ƒ Demonstrated examples of partnerships with traditional
leaders, other CBOs, other agencies, government
institutions
A second measure of decentralization is the ratio of seed distributed by the
project and by the communities. As illustrated in Figure 36 below, the amount
of seed distributed by CARE has steadily declined, while the amount of seed
distributed by AMCs has significantly increased. Data from the CSM book
indicates that the amount of seed being purchased by farmers has also
increased, meaning that the beneficiaries are becoming much less dependent
on the project for seed. Training on management of the seed scheme and
technologies for seed production and storage contribute to this capacity (see
Figure 37).
Quantities of seed distributed by CARE and AMCs
0
5
10
15
20
25
30
35
1995/96 1996/97 1997/98 1998/99
Tons
CARE AMC
Figure 36. Quantities of seed distributed by AMCs and LFSP
Source: (LFSP, 1999c)
LFSP Mid-Term Evaluation ARD
43
Figure 37. Training on constructing seed storage bins.
Training on seed storage technologies such as seed banks has helped to decentralize
responsibility for key activities.
6.1.8 Within 4 years, community institutions will exhibit independent
planning and management skills and the ability to enter into contractual
relations with other institutions.
RESULTS
INDEPENDENT CBOS: SOME EXAMPLES, BUT NOT THERE YET
A recent internal evaluation found that only 12% of AMCs are able to be in
partnership with 2-3 service providers or trainers (Hatwiinda & Whitehead,
2000). This is much less than the internal target of 80%. However the low rate
is probably more a result of the few number of institutions in the LFSP areas
than it is a measure of CBO capacity to form partnerships. The trend is
increasing though, particularly with private sector institutions where
relationships are being formed around the provision of inputs and purchasing
of crops.
There are other anecdotal examples of AMCs exhibiting independent
management and planning skills. In 1999, for example, AMCs in the
Mapatizya Constituency were involved in organizing the distribution of relief
aid. The efficiency and effectiveness of this operation was praised by Chief
Moomba and even the Deputy Minister of MAFF. Other examples of AMCs
taking initiative include mobilizing labor and resource of the construction of
dip tanks, soliciting assistance from MAFF on specific training needs, and
working with other relief agencies such as UNICEF or World Vision on
borehole projects.
LFSP Mid-Term Evaluation ARD
44
6.1.9 Within 5 years, community institutions will be exhibiting
democratic practices and full accountability to members, and will be self￾sustaining institutions
RESULTS
DEMOCRATIC AND ACCOUNTABLE CBOS: MOSTLY
SELF-SUSTAINING CBOS: STARTING
The CBOs are democratic in the sense that the leadership is elected. Although
there have been some issues with powerful people such as traditional
authorities leading autocratically, these cases seem to be the exception and not
the rule (Milimo et al., 1997). According to a recent internal evaluation, 62%
of AMC have written constitutions (Hatwiinda & Whitehead, 2000). Many
others have verbal constitutions or bylaws.
The question of self-sustainability is a more complex issue. Certainly the ten
AMCs that were selected for graduation in 1998 met the criteria that the
project felt demonstrated that they could manage on their own (see 6.1.7). The
project has yet to make a comprehensive follow-up study of these graduated
AMCs, but preliminary indications are that they are managing the existing
activities such as the seed scheme fairly well, and some have even gone into
new activities and partnerships. The fact that the majority of AMCs and
VMCs still exist is also an indication that they value the roles and services that
the CBOs make possible, which is a very promising indicator of sustainability.
Management of water points is
responsibility assumed by many
AMCs with water projects.
Maintenance and collection of
user fees are ongoing practices
that require transparency,
accountability, and self￾sufficiency.
Figure 38. Management of water points.
LFSP Mid-Term Evaluation ARD
45
6.2 Appropriate and sustainable farming systems
6.2.1 Within 3 years, all VMCs have achieved widespread dissemination
of desired drought tolerant crop varieties
RESULTS
WIDESPREAD DISSEMINATION OF DROUGHT TOLERANT SEEDS: YES
The introduction and maintenance of drought resistant crop varieties remains
the centerpiece of LFSP's contribution to improved food security in Southern
Province. Within 3 years, nearly all VMCs had achieved widespread
dissemination of 17-20 adapted varieties of 11 food crops within the project
area of Kalomo and Kazungula Districts. The only AMCs which may not have
achieved widespread dissemination of these varieties, are those where farming
is secondary to other livelihoods, such as Songwe, where ornamental carving
(curios) is economically more important than farming for many rural residents.
Other AMCs may have become dysfunctional during the first three years,
before widespread dissemination of drought tolerant varieties could occur.
Figure 39. Maize field of drought tolerant MMV400, Makunka, 1996
Figure 29 on page 30 shows the number of project beneficiaries, virtually all
of whom benefit from the seed scheme. Figure 43 on page 50 illustrates the
scope of crop diversification in the project by showing the total number of
crops and varieties distributed by LFSP per year. Table 2 below lists the
amount of seed distributed by LFSP to farmers. Once this seed reaches the
farmers, it is further multiplied and disseminated both within the project area
as well as to non-participants.
LFSP Mid-Term Evaluation ARD
46
Seed Distributed by LFSP
(all units in kg unless otherwise noted)
Crop 1994/95 1995/96 1996/97 1997/98 1998/99
Maize 12,876 14,897
MMV400 8,250 3,600 2,500
GV 412 (hybrid) 1,480
Pool 16 600 5,315 2,490
Total maize: 10,330 8,915 12,876 19,887
Sorghum 2,891 1,983
Kuyuma 678 3,962 3,232 194 3,555
Sima 9,123
Total sorghum: 678 13,085 3,232 3,085 5,538
Pearl millet 56 461
Kaufela 1,580 44
Lubasi 135 234 200
Total Pearl millet: 1,715 278 56 661
Groundnuts 3,810 1,305
Natal Common 4,748 5,005 230 6,950
Chipego 120 67
Falcon 3,000
Total Groundnuts: 4,868 5,072 4,040 11,255
Cowpea 1,425 940
Lutemmbwe 150 1,432 850 542 650
Bubebe 128 400 403 656
Total cowpea: 278 1,432 1,250 2,370 2,246
Green Gram
RSA 1 169 196 43
Total Green Gram: 169 196 43
Bambara nuts 475
red buddy 500 920
brown breed 493 921
freacle 500 920
Total bambara nuts: 1,493 3,236
Sunflower
G100 390 103
Record 2,425
Total sunflower: 390 103 2,425
Cassava (cuttings)
Bangweulu 3,171 30 1,530 2,140
Sweet Potatoes
Chingovwa 1,080
Zambezi 6,840
LFSP Mid-Term Evaluation ARD
47
Crop 1994/95 1995/96 1996/97 1997/98 1998/99
Total Sweet potatoes: 7,920
Tree Species (seedlings)
Cedrella 100
Neem tree 950
Guava 148 3
Pawpaw 174
Mulberry 165
Mango 58
Nsombo 83
Soil Improvement Crops
Sun hemp 25 180
Velvet Beans 36 130 270
Sesbania sesban 14 20
Tephrosia 6
Table 2. Amount of seed distributed by LFSP to participating farmers.
Figure 40 below shows the actual average amount of seed planted per farmer
from each of the different seed sources. It can be noted that since 1998 the
amount of CARE seed planted by farmers has been roughly constant, while
the amount of seed kept or purchased seems to be increasing. This suggests
that the relative importance of CARE's infusion of new seed into its
beneficiaries is going down as local seed stocks and private sector sources
become stronger.
CSM - Average Amount of Seed Planted by Source
0
10
20
30
40
50
60
70
80
1996 1997 1998 1999 2000
avg seed planted in k
g
AMC seed bank CARE kept purchased
Figure 40. Average amount of seed planted per household by seed source.
Source: CSM database.
LFSP Mid-Term Evaluation ARD
48
6.2.2 Within 3 years, farming system improvement experimental
activities will be in process in 80% of AMCs.
RESULTS
EXPERIMENTAL ACTIVITIES: BELOW TARGET
Currently, about 24 of the 54 AMCs are conducting demonstrations of
improved farming system technologies (see Figure 41). This represents about
45% of the AMCs, less than the 80% target by year three. A demonstration is
defined as when the Farming Systems Technical Unit provides demonstration
materials and instruction to a farmer in the AMC, usually the facilitator, who
then sets up a demonstration plot for the new technology.
Livingstone Food Security Project
Percent of AMCs with Demonstration Projects
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
1996/7 1997/8 1998/9
0
10
20
30
40
50
60
Percent AMCs with demos Target AMCs
Figure 41. Percent of AMCs with demonstration farming systems projects.
Source: LFSP Farming Systems Section.
6.2.3 Within 1 year, 30 farmer extension facilitators will have been
selected by communities, and will be operational. (60 after 3 years)
RESULTS
NUMBER OF FACILITATORS: ABOVE TARGET
Facilitators are elected by their communities to act as local extension agents.
Facilitators serve as an important entry point for extension programs, and
work with the heads of cell groups to disseminate information on activities and
new technologies. Project extension officers rely on facilitators to help with
monitoring and coordination of project activities, and have accordingly
encouraged AMCs to elect and support facilitators. There are two to three
LFSP Mid-Term Evaluation ARD
49
facilitators per AMC. The project is well above their target of 60 trained
facilitators within three years (see Figure 42).
Livingstone Food Security Project
Number of facilitators
0
10
20
30
40
50
60
70
80
90
100
1996/97 1997/98 1998/99
Actual Target
Figure 42. Number of facilitators, actual and target.
Source: LFSP.
6.2.4 Within 3 years, a farmer-to-farmer extension system will be in place
with facilitators receiving and passing information, conducting training
and facilitating the activities of the VMCs and AMCs
RESULTS
FARMER-TO-FARMER SYSTEM: IN PLACE
As noted in 6.2.3 above, the project is above target in establishing and
supporting local facilitators in almost all AMCs. The evaluation team visited
many of the facilitators during field visits, and was in general impressed with
their abilities and achievements. They frequently seemed younger, quicker,
more energetic than their counterpart AMC Chairmen. They are often the most
innovative in the community, and anticipate and challenge risk associated with
trials of new technologies.
LFSP has supported the community facilitators with an on-going training
program. Table 3 below lists the training of trainers courses which have
targeted facilitators since 1996.
Year Focus of Training Participants
male female
1996 Facilitation skills 24 0
1998 Facilitation skills 34 1
1999 Participatory Interest Group Training 47 5
1998 NRM around dams 18 0
1999 Soil moisture conservation 90 10
Table 3. Training of trainer programs held since 1996 which included facilitators in the target
audience.
LFSP Mid-Term Evaluation ARD
50
6.2.5 Within 5 years, 30% of farmers in seed groups will have adopted
one or more farming system improvement practices
RESULTS
HOUSEHOLD ADOPTION OF IMPROVED FARMING PRACTICES: DATA
NOT AVAILABLE
The Farming Systems section has worked closely with selected farmers and
CBO facilitators to setup a network of demonstrations of improved farming
practices. See the Technology Matrix in Annex V for a complete list of the
improved farming practices being promoted in LFSP.
Presently there are demonstrations in 24 out of 54 AMCs (44%). However not
all farmers in these AMCs have adopted the demonstrated technology, and the
project does not collect data on individual farmer adoption rates. There are,
however, many case studies in quarterly and technical reports of how the
improved farming system practices are improving agricultural production. In
the case of use of improved seed varieties, the adoption rate is near 100%.
6.2.6 Production of increased range of cereal, legume, cucurbit, vegetable
and fruit crops by different households
RESULTS
INCREASED RANGE OF CROPS: YES
Although production is not systemically monitored in the project, we can use
seed distribution as a reasonable proxy for production of a greater variety of
crops. Figure 43 below shows the number of crops and varieties distributed by
the project per year.
Livingstone Food Security Project
Number of Crops and Varieties Distributed
0
5
10
15
20
25
1994/95 1995/96 1996/97 1997/98 1998/99 1999/00
num crops num varieties
Figure 43. Number of crops and varieties distributed by LFSP.
Source: LFSP Farming Systems section.
LFSP Mid-Term Evaluation ARD
51
6.2.7 Improved tillage, cropping, soil fertility management and moisture
conservation practices
RESULTS
IMPROVED FARMING METHODS: YES, BUT MOSTLY IN
DEMONSTRATION PHASE
As noted in 6.2.2 above, demonstrations of improved farming practices are
currently being conducted in 24 AMCs. The Technology Matrix in Annex V
lists the specific farming methods being tested by the project.
The evaluation team encountered farmers in Kalomo using cover crops,
however because the technologies are still mostly in the demonstration phase,
there is little data on adoption rates. One of the challenges facing the project is
to develop strategies to scale-up the use of specific farming methods, and
establish methods to monitor their adoption.
Figure 44. Shifting kraal, Delevu.
Shifting kraals are being promoted as a technology to improve soil fertility in areas with
livestock with no additional labor input required by the farmer. Source: Robby Mwiinga
LFSP Mid-Term Evaluation ARD
52
6.3 Improved water harvesting and natural resource management
practices
6.3.1 Within 3 years, all AMCs will have undertaken water resource
analyses, developed priorities and action plans and are undertaking water
harvesting and supply activities
RESULTS
WATER RESOURCE ANALYSES AND PROJECT: GOOD, BUT BELOW
100% TARGET
A recent study found that 82% of the water committees were able to
independently plan, develop priorities, manage and operate water structures
(Hatwiinda & Whitehead, 2000). This is below the 100% target specified in
the Cooperative Agreement, but still impressive. Table 4 below list details on
the water projects undertaken in Kazungula since 1995, while Figure 45 shows
the location of dams. Note the water projects in Kalomo which are financed by
DFID or in partnership with UNICEF or World Vision are not included.
Dam Period with
water
Capacity
(M3)
Number
of cattle
Benefic￾iaries
Utilization
1.Delevu New dam ? 37,500 900 400 L, D
2 Siankali 12 months 25,000 700 700 L, D
3 Muzumbwe 12 months 13,000 1,000 2,000 L, D, I
4 Katapazi A New dam ? 4,200 N/A 100 I
5 Katapazi B New dam ? 4,400 N/A 700 I
6 Sinde weir. 12 months 6,000 1,000 800 FF, I, D, L
7 Jack Mwanapapa 8 months 7,220 64 150 FF, I, D, L
8 Siafwipa 12 months 7,500 900 600 L, I
9 Kooma mooka 8 months 6,000 800 300 L
10 Mapanda 9 months 7,500 558 719 L
11 Malimba 12 months 18,000 2,100 2,660 L, FF, I, D
12 Sibandwe 8 months 6,000 500 600 L
13 Zulu weir 11 months 6,500 300 2,500 L, D
Total 148,820 8,822 12,229
Table 4. Dams constructed in Kazungula district since 1995.
Source: (LFSP, 1999a). Utilization codes: FF = fish farming; L = livestock watering
reservoirs; I = irrigation and vegetable gardening; D = domestic reservoir for human
consumption.
LFSP Mid-Term Evaluation ARD
53
Figure 45. Location of dams built in Kazungula since 1995.
Water harvesting
Structures
Structures
Completed
Rehabs New Cattle
Population
Irrigated
area (Ha)
# of
Beneficiaries
Boreholes 47 7 40 12,000
Dams & weirs 13 7 6 8,722 20 10,270
Wells 20 0 20 5 20
Springs 7 7 0 1,750
Artisan wells 3 3 3 300
Irrigation pumps 1 0 1 3 200
Total 84 23 70 8,722 25 24,540
Table 5. Number/type of water harvesting structures in Kazungula, 1995-1998.
Figure 46. Siabandwa dam undergoing rehabilitation with tractor and dam scoop.
Source: Robby Mwiinga.
LFSP Mid-Term Evaluation ARD
54
Figure 47. Weir rehabilitation in Nyawa.
About 1/4 of dam/weir projects are rehabilitations like this one.
6.3.2 All water committees are affiliated with district WASHE structure.
RESULTS
WATER COMMITTEES AFFILIATED WITH WASHE: YES
By the end of the third year of the project, all 63 Water Committees were
affiliated with D-WASHE structures in both Kalomo and Kazungula Districts.
6.3.3 Establishment of dry season irrigation and market gardening
schemes in 5 AMCs within 3 years
RESULTS
DRY SEASON IRRIGATION AND MARKET GARDENING: TARGET MET
Dry season irrigation schemes for vegetable and fruit production activities
have been established in 7 AMCs, resulting from dam and weir construction.
Dry season vegetable production activities have also been initiated in 5 as a
result of well construction activities of the project. Twenty hectares are
currently being irrigated from the 7 dam and weir constructions and 5 hectares
are being irrigated from well constructions. Table 10 below shows that there
are 7,010 AMC/VMC members who are beneficiaries of vegetable garden
plots, irrigated from dams and weirs (although the average 29 m2
 per
LFSP Mid-Term Evaluation ARD
55
beneficiary is a very small garden), and 20 beneficiaries of vegetable
gardening plots, irrigated from wells.
New Water Source Vegetable
Production Area
Vegetable Production
Beneficiaries
Dams & weirs 20 ha 7,010
Wells 5 ha 20
Table 6. Vegetable production resulting from water harvesting technologies.
Source: LFSP.
The SEAD section has also been
involved in boosting dry season
vegetable production through the
promotion of treadle pumps. The
project has given out 19 treadle
pumps on a loan basis for
demonstration purposes. To supply
the pumps LFSP partners with
International Development
Enterprises, a Lusaka-based NGO
which provides a cheaper source of
locally manufactured pumps.
After the pumps have been
demonstrated in an area, farmers
who with to obtain one are required
to purchase them directly from the
supplier. To build up rapport
between the farmers and the supplier
of treadle pumps, LFSP serves only
as a coordinator and credit provider
for the demonstrations.
Figure 48. Woman demonstrating the treadle
pump for vegetable irrigation
 The supplier conducts the actual installation and training of the pumps.
Although no farmers have yet purchased pumps after seeing the
demonstrations, several have made inquiries.
6.3.4 Experimentation with on-farm water harvesting practices by
farmers in 50% of water committee areas within 3 years.
RESULTS
EXPERIMENTATION WITH ON-FARM WATER HARVESTING:
EXPERIMENTATION IS TAKING PLACE, BUT QUANTITATIVE DATA NOT
AVAILALBLE
Although there are a lot of case studies on how water projects are benefiting
LFSP farmers (see box below for an example), the evaluation team was not
able to find a synthesis on the number of farmers experimenting with on-farm
water harvesting practices.
LFSP Mid-Term Evaluation ARD
56
Effects of training on dam construction and NRM on Katapazi community
Mr. Jossam Sichonti of Katapazi situated 60km north of Livingstone has
constructed an embankment of about 2m high, to harvest about 1200m3
 of
water from a catchment of about 3km2
. He has put up a 1½” pipe at the base of
the embankment in order to drain off excess water downstream where he
intends to utilize the same water for irrigating vegetables.
On the upstream part of the dam. He has ploughed 3500m2
 of land where he is
going to grow leguminous crops and maize. He has put up gabions as part of
contour ridging and he intends to trap most of the rich silt in order to improve
his soil fertility.
The following are the benefits of his embankment which he cites:
ƒ Improved soil fertility by deposition of silt and increase in the soil
moisture content.
ƒ Irrigating his vegetables and other useful plants like vetiver grass.
ƒ Increasing the water table for future use (e.g., to construct shallow wells
for irrigation purposes).
ƒ To resuscitate the springs and to increase their discharge by recharging the
underground water table.
ƒ To plant vetiver grass as part of contour ridging for long term
sustainability.
Lastly, he thanked CARE for teaching him on how to manage the natural
resources and different methods of harvesting the water.
6.3.5 Within 3 years, 30% of AMCs have developed and are
implementing broader resource management guidelines and strategies
RESULTS
BROADER RESOURCE MANAGEMENT PRACTICSES: ABOVE TARGET
A recent internal evaluation found that 80% of the AMCs are currently
implementing broader NRM guidelines and strategies (Hatwiinda &
Whitehead, 2000). NRM strategies include use of technologies such as vetiver
grass of stabilizing banks around weirs and dams, zoning around water points
for human and animal use, and other soil erosion mechanisms such as gabions
and stonelines to trap silt and prevent erosions around gullies.
The high adoption of resource management practices around water projects
should not be surprising, as many of the technologies are designed to reduce
siltation which is widely acknowledged to be one of the factors than can
reduce the length of time during which standing water remains after the rains.
Erosion prevention technologies are therefore highly in demand and part and
parcel of most water structure rehabilitation or construction projects.
LFSP Mid-Term Evaluation ARD
57
However the project should not rest on its laurels in terms of NRM adoption,
as there is still much unmet need for NRM practices. Most of the NRM
practices are focused on the prevention of siltation in the immediate vicinity of
water points. The water section acknowledges the need for broader watershed
management practices, however this has been more difficult to sell to the
water committees because the benefits are not immediately seen. The project
recognizes it has more work to do in demonstrating the long-term benefits of
NRM practices on water quantity and quality. To move in this direction, it
needs to focus on more participatory monitoring programs using tools such as
erosion pins, vegetation sampling, water quality monitoring, and field
mapping.
Figure 49. A water committee visits Siabandwa dam.
Exchange visits help communities learn about dam construction, maintenance, and the role of
NRM in preventing siltation. Here newly planted vetiver grass can be seen on the dam wall.
Source: Robby Mwiinga.
LFSP Mid-Term Evaluation ARD
58
6.4 Increased incomes and income earning opportunities
6.4.1 Analyses undertaken of household livelihood strategies
RESULTS
LIVELIHOOD STRATEGIES ANALYZED: COMPLETE
Household livelihood strategies are central to LFSP's operational strategy, and
have been the focus of several internal and external reviews. They were a
major focus of the of the PRA exercises in 1995 that formed the basis of the
project intervention strategy. They were also the focus of a marketing
consultancy in 1997 (Milimo et al., 1997). More recently, the 1999 internal
mid-term review reviewed livelihood strategies using PRA techniques such as
food calendars (Ndiyoi et al., 1999).
The project continues to analyze livelihood strategies whenever new PRAs are
conducted including in the recent expansion areas. The Food Production
Trends Survey and Community Self-Monitoring books provide additional
insights into the strategies used by rural households in terms of crop mix,
coping strategies, and land use.
All of the above mentioned data collection methods generate useful
information, but a complete understanding of why people do what they do (or
don’t do what they don’t do) is not yet a battle won. There are several issues
that need to be further explored, because the success of new action lines such
as livestock management and NRM depend on a sound understanding of
livelihood requirements and options.
6.4.2 Diversification and expansion of a range of income earning
opportunities
RESULTS
DIVERSIFICATION OF INCOME EARNING OPPORTUNITIES: YES, BUT ON
A LIMITED SCALE
LFSP's activities to diversify income-earning opportunities have focused on
combining CBO capacity building methods, appropriate technologies, and
market linkages to support the development of small group enterprises. Some
beneficiaries have capitalized on skills and technologies promoted by the
project to enter small scale market ventures on their own, such as vegetable
gardening or handicrafts, with little or no dependence on LFSP. However
operationally the income generation component of the project is focused
around Interest Groups. Interest groups fall under the AMC level and are
thematically oriented around an enterprise strategy or capacity building theme.
LFSP Mid-Term Evaluation ARD
59
Presently there are approximately 205 Interest Groups listed with the project,
with over 2,000 members (60% male, 40% female). Judging only by the titles,
some of these interest groups may only exist on paper, and it will take time to
determine which ones truly develop into viable group enterprises. Figure 50
below shows the breakdown of the different interest groups by focus.
Livingstone Food Security Project
Interest Groups by Category
81
33
21
18
14
13
13
5 4 3
Small industry
Small livestock
Vegetable production
Cattle
Field crop production
Unspecified / non-market
NRM
Post harvest processing
Water
Cooking
Figure 50. Interest groups by category.
Source: LFSP.
Small Industry Interest Group Topics
ƒ Baking
ƒ Basket making
ƒ Beer brewing
ƒ Black smithing
ƒ Brick laying
ƒ Brick making
ƒ Carpentry
ƒ Carving
ƒ Cookery
ƒ Curios
ƒ Goat rearing
ƒ Grocery projects
ƒ Household and laundry
ƒ Knitting
ƒ Mats & brooms
ƒ Natural pest control
ƒ Piggery production
ƒ Pit latrines
ƒ Pit sawing
ƒ Poultry production
ƒ Sewing
ƒ Tailoring
The project has begun to note that most of the training topics requested for by
interest groups lies outside LFSP's range of staff skills and capacity. This has
caused some delay in implementation and training (LFSP, 1999a). This
finding also highlights the need to focus on forging institutional partnerships
to better support interest groups. For example, the Export Board of Zambia has
LFSP Mid-Term Evaluation ARD
60
assisted on several occasions in training and setting up market linkages in
areas such as curios production and thatching grass.
Although LFSP doesn't monitor or maintain a central database of income
earning activities, there are plenty of anecdotal examples of increased rural
income through IGAs (see box below).
Mandia Vegetable Gardening
In 1996 CARE provided group loans to the 10 vegetable interest
groups. The loan was in the form of inputs; seed, chemicals and PVC
pipes. The loan amount disbursed was K373, 615.00. Group peer
pressure was to be used in recoveries. The loan was to be paid over a
period of 6 months. Full loan repayment was made in 1997. A mini
research indicated that over one million Kwacha was generated
besides full loan repayment. 75% of the participants were able to buy
school uniforms and shoes for their children. Part of the income
realized was used to buy other necessities such as groceries.
Figure 51. Thatching grass ready to be sold.
Thatching grass is an example of an income generating activity supported by LFSP. Buyers
include hotels in Livingstone and Zimbabwe. Source: LFSP.
6.4.3 Expansion of internal trading and marketing within 2 years of seed
scheme become operational in each area.
RESULTS
EXPANSION OF INTERNAL TRADING AND MARKETING: YES, BUT
LITTLE DATA TO BACK IT UP
There is strong anecdotal evidence that farmers are bartering maize and other
crops among themselves and with non-participating farmers though internal
trading networks. Cowpea is an example of one crop where in some areas
LFSP Mid-Term Evaluation ARD
61
much of the demand for local production is internal. Seed is also exchanged
and sold up to 50 km from the source (Milimo & Tripp, 1999).
Despite the seemingly common practice of internal trading and marketing, the
project has not documented or attempted to quantify the scope of the trade or
the networks used. This would be a difficult process to measure accurately,
however the project could begin to capture the scope of internal trading by
including crop sales and trading in the CSM.
6.4.4 Pilot marketing and schemes established in 5 AMCs within 1 year
RESULTS
PILOT MARKETING SCHEMES: YES (BUT NOT BY YEAR ONE)
For a discussion see 6.4.5 below.
6.4.5 Within 4 years, market linkages established between traders and
AMCs, or directly with urban centers, for 80% of the AMCs
RESULTS
LINKAGES WITH URBAN MARKETS: BELOW TARGET
During the formative PRA exercises conducted in 1995, the lack of access to
markets was identified as one of the barriers to improving food security and
raising rural income. Farmers had trouble both buying goods such as food and
inputs, and selling their produce. Hence marketing was selected as one of the
focal activities of the SEAD section.
Five years later, the need for stronger marketing has only increased. The
sustained demand is due to 1) increased production surplus due to improved
farming practices, 2) the need to raise household income as a way of
strengthening the asset base to mitigate against drought, 3) the need to build
structures and linkages that will outlast the project.
LFSP has tried to support marketing through three main mechanisms, 1)
outgrower schemes, 2) community training, 3) direct involvement in setting up
linkages.
Outgrower schemes
The first outgrower scheme was with Bimzi for paprika. The plan looked
promising in the beginning. Two AMCs, Katapazi and Siandazi, were selected
to participate, and CARE distributed 48 x 500g of paprika seed on a loan
basis. However from there things began to fall apart. Bimzi was supposed to
provide extension training and fungicide, but didn't. Farmers didn't understand
the water requirements of the plants and relied only on natural rainfall instead
of irrigating. Some never transplanted the seedlings. Of the expected 20 tonnes
of paprika only 0.5 was produced. Bimzi refused to send a truck for such a
small amount, and wanted to CARE to pick up the cost of transport. They also
stated they would only buy the highest grade. Farmers started side-selling to
LFSP Mid-Term Evaluation ARD
62
Cheetah Zambia. In short, it was a disaster. However all was not lost with the
paprika outgrower scheme, as it was tested on a small scale and valuable
lessons were learned (see box below).
Lessons Learned from the Paprika Outgrower Scheme
ƒ Ensure that the agribusiness, the CBOs, and the project have
a very clear understanding of each other's responsibilities.
ƒ Ensure that farmers are capable of growing the crop. In the
case of paprika, only farmers who have the ability to irrigate
should be allowed to participate.
ƒ A crop which has never been grown before in the area is not
the best choice for an outgrower scheme.
ƒ Start small, and cut your losses early if it doesn't work.
ƒ Credit makes everything more complicated.
ƒ Frequent extension contact is needed to ensure that farmers
are complying with the terms of the agreement.
Based on the lessons learned from the first outgrower attempt, LFSP organized
two other outgrower programs. An outgrower scheme based on maize, a crop
which is already proven and well-known to the farmers, is currently in its
second year with Omnia. This scheme has no credit component, so all inputs
have to be purchased with cash. Omnia rents sheds from the Food Reserve
Agency (FRA) in the areas of the scheme, so transport is less of an issue. In
addition to the Omnia outgrower scheme, there are also new outgrower
schemes with Castor Company for castor, and Amanita for sunflower and
sorghum. Figure 52 shows the locations of the outgrower schemes.
Amanita - sunflower
Amanita - sorghum
Castor Company - castor
Omnia - maize
Livingstone
Zimba
Kalomo
Figure 52. Locations of outgrower schemes
LFSP Mid-Term Evaluation ARD
63
Community Training
Although training in marketing was initiated fairly late, it has shown some
promising results. Market training was conducted for about 40 farmers in
Nyawa and Mweemba in March 1999. As a result, farmers became more
oriented to market conditions in Zambia, and Omnia started selling inputs and
buying maize from LFSP farmers to the tune of K75,000,000 in 1999. In April
1999, another marketing training was conducted in Makunka for about 35
farmers. This also resulted in Omnia renting a shed from the FRA to begin
selling inputs and buying maize from farmers.
The project has also facilitated marketing along the 'vegetable belt' in the
escarpment area, Kasiya, Jack, and Mandandi. LFSP staff helped vegetable
growers there set up a market information system, a summary of vegetable
varieties, volumes, and availability dates which was distributed to traders in
Livingstone. This resulted in a number of orders from traders as far away as
Botswana and Namibia.
A topical appraisal in forest product marketing was carried out in Chief
Sekute's area in 1999. This revealed that a significant amount of income was
being generated from forest products in the form of timber and reed mats.
Results from this appraisal were incorporated into a proposal on Joint Forestry
Management.
Direct involvement in setting up linkages
The project is in the process of playing a direct negotiating role in setting up
market linkages with other private sector bodies, including Finta (milk), the
Export Board of Zambia (curios, vegetables), Sun International (thatching
grass), and National Milling (sorghum). These linkages were first identified in
1998 (LFSP, 1998b), however have taken a long time to develop.
Summary
LFSP has made some promising inroads into establishing external market
linkages through outgrower schemes, training, market information systems,
and direct facilitation. This has resulted in new market linkages for vegetables,
maize, castor, sunflower, sorghum, curios, grass, and miscellaneous forest
products.
However the benefits of the marketing efforts have reached only a fraction of
the AMCs, and although the data is not well organized, one can safely
conclude that they have fallen short of the target of reaching 80% of AMCs.
The achievements are generally isolated, small scale, and labor intensive. A
reasonable question to ask is why has the marketing efforts progressed so
slowly?
There are several challenges that have faced LFSP's efforts to establish market
linkages, and will probably continue to do so. First, agricultural production
LFSP Mid-Term Evaluation ARD
64
levels are improving but still relatively low1
 relative to other areas that
compete in the market economy. Second, many of the LFSP farmers are far
away from the road network, making communications and transactions
logistically challenging and expensive. Third, LFSP has had difficulty finding
a qualified staff person to head the SEAD section, and the position is currently
vacant. Fourth, unlike the farming systems section, the farmer-to-farmer
network has not been effectively used to spread marketing skills. And finally,
developing marketing linkages where none exist simply takes a lot of work
and time.
It may also be that much of the achievements in marketing simply haven't
been captured by the project. Monitoring and documentation of the SEAD
section is probably the weakest in the project. There are a lot of case studies
and examples of individual AMCs, but the information is not well organized
and the SEAD section has not developed a standardized reporting format for
impact monitoring. Unlike the seed scheme or water section, there is no master
database of all beneficiaries of marketing activities. This is partly due to the
vacancy of the SEAD coordinator position, and partly due to the difficulty of
capturing information on income generation and informal trade. One way that
trading can be better monitored is to record it in the CSM.
However there are already some lessons to be learned about the strategies that
LFSP has so far used to strengthen market linkages. One of the objectives of a
marketing program should be to maximize cost-effectiveness, in other words
setting up the greatest number of market linkages for the least amount of cost.
In terms of cost-effectiveness, some of the strategies used by LFSP have been
more cost efficient than others. Outgrower schemes are fairly time-intensive to
set up, however they have the potential of reaching a fair number of farmers
and once established they can grow on their own. The outgrower scheme with
Omnia which focuses on maize production, has the potential of growing to
reach a greater number of farmers. Maize production is well known among
farmers, so the main limiting factors with this scheme are the capacity and
interest of Omnia, the availability of sheds to rent, and production volume.
On the other hand, a strategy based on organizing workshops and using project
staff to contact traders directly to market forest products, curios, or thatching
grass may result in new market linkages on a limited scale, but may not be a
strategy that will likely have broad impact or be easily replicated elsewhere.
This approach of direct support of small-scale marketing may be the best that
can be done in the difficult environment, but there are some other strategies
that seem to have yielded broader results for less investment. The development
of market information systems has worked particularly well for vegetable
growers in the escarpment zone. If the method can be simplified and made less
dependent on LFSP staff, then this could be a technology that could be used
on a wider basis. This is a marketing technology which is also easily
transferable to other commodities such as forest products.
 
1
 The one exception is vegetable production in the escarpment, which benefits both from adequate
water and also close proximity to Livingstone.
LFSP Mid-Term Evaluation ARD
65
Other strategies that may help broaden the reach of LFSP's marketing efforts
include forming partnerships with institutions such as MAFF extension,
whereby MAFF officers can be trained to teach marketing strategies and
spread marketing information. There may be other institutions with a presence
in the project area that could help serve as a conduit for marketing efforts.
Something comparable to the farmer-to-farmer system, which has worked so
well for the farming systems unit, may also help increase the impact of the
marketing program.
Ultimately, the bad roads, poor soils, and marginal rainfall are going to limit
the effectiveness of any approach to build market linkages, however the
project should concentrate on strategies that will have the broadest impact, can
be replicated in other areas, and sustained past the project completion date.
6.4.6 Saving schemes established with 4 AMCs in 2 years and 10 AMCs in
4 years (approximately 1,000 then 3,000 member HHs)
6.4.7 Savings and credit schemes established within 2 AMCs in 2 years
(approx. 500 HHs) and 5 AMCs in 4 years (approx. 500 then 1,500
member HHs)
6.4.8 Management and monitoring systems established for savings and
credit schemes
6.4.9 All members of schemes trained and understand scheme systems
6.4.10 Management of schemes undertaken by elected committees
RESULTS
SAVINGS AND CREDIT SCHEME: DIDN'T WORK
During the 1995 PRA exercises, it was learnt that people save in the form of
livestock, (cattle and other small livestock), grain, and in the form of plain
cash. It was further learnt that people find it difficult to open bank accounts
because of the high minimum balances, monthly fees, and distance to banks.
Subsequently, the project developed a pilot savings and loan scheme model
and incorporated it into the design of LSFP. The rationale behind increasing
household savings included:
ƒ In times of hunger one can withdraw his savings to buy food.
ƒ When savings accumulate, it becomes easier to buy bigger assets such
as production assets like ploughs and work oxen.
ƒ Savings provides a sense of security
ƒ Borrowing from friends, relatives, lending institutions and other
becomes easier if one has savings as collateral.
LFSP Mid-Term Evaluation ARD
66
Based on the PRA results, LFSP recognized that the highest potential to
generate income was in the escarpment, where vegetable gardening is
practiced. Consequently, the savings scheme was piloted in two AMCs in the
escarpment zone, Mandia and Katapazi.
Currently, the savings scheme is running in five VMCs (Lizazi, Katapazi,
Libonda, Makamisa, and Jack) in the escarpment area of Chief Sekute and
Chief Mukuni. All the VMCs are depositing their savings with Barclays Bank
through the CARE account, and CARE absorbs the bank fees. This account
earns interest at the rate of 11.5% per annum, which is less than the 20-30%
annual inflation. It's estimated that 99% of community savings banked so far
have been generated from vegetable sales.
Training of savings group members was conducted by the project, however a
viable managing and monitoring system was not established for savings
groups. This was because of the short duration of the activity and also because
the money saved was in the CARE account in Livingstone, not in the account
of the savings groups themselves.
By 1998, savings had completely stagnated. To date a total of only $460 has
been banked with the scheme. Reasons for the low rate of savings are
summarized in the box below.
Reasons why the savings and loan scheme didn't take off
ƒ The loan scheme was modeled on a scheme from the PULSE
program, which is based in an urban area where income is
higher and more regular
ƒ Income in rural areas is seasonal, making regular savings
difficult
ƒ Income in most LFSP areas is marginal to begin with, and often
needed to buy food for the hungry season
ƒ The bank interest is lower than the inflation rate, so farmers lose
value on their money rather than gain
ƒ Savings was linked to loans, so the incentive to deposit money
was not to save for a rainy day but to leverage project loans
ƒ Loans were delayed or never materialized, so people became
discouraged
Even if some of the above problems hadn't occurred, there were flaws in the
design of the savings scheme. The only real incentive to save was the hope of
getting a loan. A few people got loans, however many didn't and lost their
interest in the scheme. The management of the accounts was also run by
CARE. Thus farmers did not get involved in managing their accounts, so it
would not have been very sustainable even if the bank charges and high
inflation didn't reduce the value of the savings.
LFSP states they will soon drop the savings scheme entirely. They have
written to the bank to give the exact interest rates so they know how much to
LFSP Mid-Term Evaluation ARD
67
return to the farmers. It was a good try and lessons were learned, but the loan
scheme did not work very well.
6.5 Reduced vulnerability to drought and improved household food
security
6.5.1 Within 5 years, 80% of households which are members of seed
groups are exhibiting improved food and livelihood security and reduced
vulnerability to drought.
RESULTS
IMPROVED FOOD SECURITY: QUALITATIVELY: DEFINITELY YES
QUANTITATIVELY: PROBABLY YES
The ultimate goal of LFSP is to improve household food security and reduce
vulnerability to drought. These are both complex variable to measure and can
best be assessed by looking for agreement in a variety of indicators.
The internal mid-term review, which conducted rapid rural appraisals in ten
AMC centers over a two-week period in June 1999, revisited many of the
same issues as the original PRAs held in 1995. They noted a significant
increase in the amount of locally produced food in the diet of the LFSP
communities, from 5% in 1995 to 41%. On the other hand, relief food, which
was a major coping strategy in the first half of the decade, contributed only
5% to household food budgets in 1998 (see Figure 53). The appraisals also
found that 34% of the households are now maize secure over a period of 9
months, compared to only 5% having food for less than six months in 1995
(Ndiyoi et al., 1999).
Livingstone Food Security Project
Household Food Budget, 1998
Barter
20%
Own production
41%
Piecework
14%
Purchase
20%
Relief
5%
Figure 53. Household food sources, 1998.
Source: (Ndiyoi et al., 1999).
LFSP Mid-Term Evaluation ARD
68
The CSM database provides other measures of project impact on food
security. Figure 54 shows slight increases in the post-harvest period with food
for maize for the escarpment2
, kalahari, and plateau. Although somewhat
preliminary due to the small sample sizes (see Figure 34 on page 38), when
the project Monitoring and Evaluation team enters more data into the CSM
database this measure of impact on food security will become more definitive.
Studies by the project have also detected a slight increase in overall food
availability (see Figure 54).
CSM: Average Period with Food - Maize
0
2
4
6
8
10
escarpment kalahari plateau
months
1996 1998 1999 2000
Figure 54. Average period with food for maize.
Source: CSM database.
Post-Harvest Food Availability
Maize
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
At least 4 At least 7 At least 10 More than 10
Number of months with food
Percentage farmers
1997 1998 1999
Figure 55. Post-harvest period with food, maize.
Source: LFSP, Quarterly Report # 14, April 2000.
 
2
 the 1997 data for the escarpment is based on only 3 households and should be disregarded
LFSP Mid-Term Evaluation ARD
69
One of the measures of reduced vulnerability to drought is an increased asset
base. Households can use assets to increase their means of production or sell
assets to buy food in times of shortage. The CSM ledgers record household
assets. As can be seen in Figure 56, there is no strong trend either up or down
in the average total value of household assets for any of the agroecological
zones. This result might be a product of the relatively small sample sizes for
the earlier years, variance in the types of assets recorded in the CSM, impacts
on the asset base including corridor disease, or simply no strong impact.
Figure 57 presents the same analysis but with only non-cattle production
assets. This restriction should eliminate the effects of inconsistent recording
procedures, because all CSMs record at least the minimum production assets,
and also eliminate any effects of corridor disease. However once again there is
no strong trend visible in the asset base. Finally, Figure 58 tracks the average
value of all household assets for the same set of households who have multiple
years worth of data. Again there is no compelling trend evident for only two
years of data. After the M&E section enters more CSM records into the
database, these charts can be reproduced and the impact of the project on the
asset base reevaluated for the final evaluation.
CSM - Average Value of Total Household Assets
$0
$200
$400
$600
$800
$1,000
$1,200
escarpment kalahari plateau
1996 1998 1999 2000
Figure 56. Average value of all household assets.
Source: CSM database.
LFSP Mid-Term Evaluation ARD
70
CSM - Average Value of Total Household Assets
Production Assets Minus Cattle
$0
$50
$100
$150
$200
$250
$300
escarpment kalahari plateau
1996 1997 1998 1999 2000
Figure 57. Average value of all production assets minus cattle.
Source: CSM database.
CSM - Tracking the Average Value of Assets 
for the Same Set of Households
$0
$100
$200
$300
$400
$500
$600
$700
escarpment kalahari plateau
Avg Total Asset 
Value
0
5
10
15
20
Num Households
1998 1999 NumHH
Figure 58. Average total value of assets for the same set of households.
Source: CSM database.
With the exception of the CSM and Food Production Trends Survey, the
project does not collect real data on production yields, which would be useful
for measuring the impact of the new seed varieties and farming methods.
Other data which is either lacking or disorganized include figures on income
generation and the impact of water projects on production and food stocks.
However qualitatively there appears to be consensus among project staff that
where IGAs, water projects, and demonstrations of improved farming
practices are underway, production and household food stocks are improving.
7.0 GENDER
The LFSP recognizes both in design and practice the important role women play in
food production, poverty alleviation, and rural development. One of the strengths of
participatory development methods is a stronger integration of the interests and needs
of women.
LFSP Mid-Term Evaluation ARD
71
Efforts to address the needs of women include a gender workshop held in January
1998 to identify possible areas of collaboration and partnership with a number of
institutions and enhance LFSP gender intervention strategies at the institution and
community level (LFSP, 1998c). Other gender sensitivity workshops have been held
for CBOs, and the project has made efforts to encourage more women to take
leadership roles in AMCs. The project also does an excellent job at monitoring the
participation of women in the project. Virtually every dataset on participation,
training, production, etc. is disaggregated by gender.
Although there are no performance indicators in the cooperative agreement that are
gender specific, the project has developed its own indicators on gender. A recent
review found that 89% of women in VMCs participate in project activities including
the seed scheme (see Figure 59), which is substantially higher than the internal target
of 40% (Hatwiinda & Whitehead, 2000). Furthermore, 37% of leadership positions in
CBOs are held by women, which is greater than the target of 30% (see Figure 60).
The same study found that 86% of women participate in programming activities for
water harvesting and utilization projects, greater than the target of 40%.
Although the number of women in leadership positions and general membership has
numerically increased, there remain barriers to the full participation of women in
CBOs. Previous studies, (e.g., Milimo et al., 1997), found that the traditional culture
in Tonga society continues to regard women as subordinate to men, and this is
reflected in their lower participation and attendance at meetings. Girl education,
which has been shown to empower women, is not very strong in the project area and
literacy remains a problem for women engaged in small group businesses. Increasing
the income and food security of women is helping, but it will certainly take a long
time for the social status of women to improve significantly.
Livingstone Food Security Project 
Gender Representation in AMC General Membership
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1997 1998 1999
Males Females
Figure 59. Gender representation in CBO general membership.
Source: (Hatwiinda & Whitehead, 2000).
LFSP Mid-Term Evaluation ARD
72
Livingstone Food Security Project 
Gender Representation in AMC Leadership
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1997 1998 1999
Males Females
Figure 60 - Gender representation in CBO leadership.
Source: (Hatwiinda & Whitehead, 2000).
Women are integrated into all activity areas in LFSP, however they form the primary
audience for several specific crops and income generating activities. Women are
traditionally responsible for the provision of relish, however are also gaining access to
men's crops. As can be seen in Figure 61 below, women plant the majority of
groundnuts, sorghum, bambara nuts, and millet. Cowpeas, which is traditionally
known as a women's crop, is increasing demanded by men due to an unfolding market
both locally and in town. The promotion of these crops by LFSP, through the
introduction of improved seed varieties, is one of the methods used by the project to
empower women because it raises their income. The only crop where women are
significantly under-represented is maize.
In terms of income generating activities, women are the primary participants in
sorghum beer making, which has been the focus of pilot training in business skills.
Women also constitute the majority of vegetable producers, which has been perhaps
the most successful example of improved marketing.
LFSP Mid-Term Evaluation ARD
73
Livingstone Food Security Project
Proportion of Men and Women Cultivating Different Crops in 1998
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Ground
Nuts
Cowpeas Sorghum Maize Bambara Millet
Female Male
Figure 61. Proportion of Men and Women Cultivating Different Crops in 1998.
Source: (LFSP, 1999a).
LFSP can continue to serve the special needs of women by monitoring their
participation in project activities and providing services to support women's
enterprises. Women in particular are in need of business skills, particularly as they get
more involved in important market-based enterprises such as vegetable production.
LFSP should also remember the needs of women when planning its future
programming, including developing partnerships with local service providing
institutions. The high growth rate of the district (est. 5% per annum) and heavily
skewed age structure of beneficiary households (see Figure 27 on page 29) suggest a
strong need for family planning services for women. Basic literacy is likewise outside
the mandate of LFSP, but definitely a need of rural women who receive less formal
education than men.
LFSP Mid-Term Evaluation ARD
74
8.0 INSTITUTIONAL RELATIONSHIPS
LFSP has sought to develop multiple partnerships with government departments,
other NGOs, and private sector institutions that have a vested interest or mandate
related to food production or the welfare of rural people. The relationships with some
of its main institutional partners are described below.
8.1 Ministry of Agriculture, Food and Fisheries (MAFF)
LFSP considers MAFF to be one of its major collaborative partners in large part
because of their common focus on agriculture. Results of the partnership can be
seen in a number of collaborative activities in which MAFF staff have participated
at provincial, district and community (camp) levels. The collaboration has
involved the departments of Field Services, Research and Specialist Services, and
the Seed Control and Certification Institute (SCCI).
MAFF staff at district level participated in the initial PRAs, which helped shape
the design of the project activities. In 1997 a partnership workshop involving
MAFF was held and was aimed at defining roles and responsibilities of each
partner. Drafting of an MOU was initiated around that time, but was never
followed up.
Other examples of collaboration with MAFF include the participation of MAFF
officers as resource persons in training. This has been the case especially in the
areas of livestock health management, food processing, and to a certain extent
crop production. LFSP has on several occasions sought advice from MAFF staff
and used them to monitor crop performance in the field. Other collaborative
activities include the testing of disease resistance vegetable varieties for
adaptability and marketability. This activity was conducted in collaboration with
the vegetable research program of the Soils and Crops Research Branch (SCRB).
CARE has provided training in participatory approaches to MAFF staff at district
and provincial levels. This resulted in the adoption of some of the aspects of the
LFSP participatory extension methods. For example an enthusiastic MAFF officer
in Gwembe district began forming VMCs after being exposed to CARE's
extension approach. (See section 10.1 on page 84 for a comparison of the LFSP
and MAFF extension strategies.)
LFSP and MAFF at district level worked together on the modalities of graduating
ten AMCs in 1998. This process involved the phased withdrawal of LFSP staff,
and the increased involvement of MAFF camp officers in five of the ten AMCs.
To facilitate a smooth transfer, the DACO shifted stronger officers into the
graduated areas, while CARE provided intensive training on participatory
approaches to extension. The graduation process seems to have worked well in a
few AMCs, including Mandia, Msokoswtani, and Mukuni, however has stagnated
in others.
LFSP Mid-Term Evaluation ARD
75
Other collaborative activities have been the participation of MAFF staff in the
project’s technical review meetings. MAFF has also been involved in constructing
dams through the Rural Investment Fund (RIF).
MAFF and LFSP have enjoyed good relationships at the national level due to
conducive policy frameworks. However there have been constraints in fruitful
partnering at the district level. One of the major problems that constrains MAFF
staff collaborating in LFSP programs is the inadequate or lack of logistics,
especially transport. The project is does not know how to handle this issue.
Another major problem, common to all government institutions, are poor salary
levels resulting in low motivation. Another source of conflict has been the GRZ
condition of service which requires payment of lunch allowance for staff in the
field. CARE policy doesn't allow lunch allowance and does not even pay sitting
allowance. See Section 6.1.3 on page 33 for additional discussion of the problems
between MAFF and LFSP.
Despite the difficulties, the partnership with MAFF has yielded some positive
results, particularly in terms of an increased appreciation by MAFF for the CBO
extension approaches promoted by LFSP.
8.2 Other Government Institutions and Agencies
Other important partners among government institutions include the Forest
Department, Social Welfare Department, National Heritage Commission and the
District Council.
The Forest Department has participated
in special PRAs to prepare for the NRM
Action Research program in Katapazi
and Delevu. This included the donation
of fruit tree seedlings for an awareness
and education event (see Figure 62).
Forestry staff have also served as
resource persons for training on
watershed management.
CARE has in turn assisted the Forest
department with training materials to in
participatory management approaches.
These materials and collaboration on
specific activities can help form a bridge
between communities and forestry, a
relationship which has been historically
chilly due to the Forestry Department's
operational focus on policing.
Figure 62. Chief Sekute plants a guava
seedling donated by the Forestry Department
Source: Robby Mwiinga, LFSP
LFSP Mid-Term Evaluation ARD
76
The LFSP water section collaborates with a number of government partners
through participation on the District Water Sanitation and Health Education
Committee (D-WASHE). Institutions represented on this committee include
MAFF, CARE, the District Council, Ministry of Health, Ministry of Community
Development and Social Services and Forestry Department.
8.3 Traditional Leaders
Traditional leaders play an important role in the rural areas of Zambia. They are
vested with powers for land allocation, and have historically played a central role
in the management of natural resources including forests and wildlife.
LFSP has also embarked on sensitization campaign of the chiefs and headmen on
the role of CBOs and their relationships with the traditional institutions. This was
meant to prevent misunderstandings which have arisen in the past.
LFSP has for the most part been able to avoid the serious problems with autocratic
traditional authorities that have plagued other NRM projects like ADMADE.
However there have been some problem areas. In one AMC, LFSP requested that
a new election be held because the chairman, how was aligned with the chief, had
misused some cement. The community was informed of this issue but the
influential chairman was reelected.
However according to project staff, problems with traditional rulers monopolizing
CBOs are the exception and not the rule. The built-in leadership functions and
democratic nature of LFSP CBOs help to place the power of traditional authorities
in balance. Written constitutions, which have been adopted by 62% of AMCs,
help to clarify the roles of all members of CBOs. LFSP is making an effort to
work more closely with traditional rulers, particularly as the project expands into
forest management and NRM practices which are traditional responsibilities of the
chief.
8.4 Private Sector
CARE/LFSP has established linkages with several private sector partners and has
carried out a number of collaborative activities. These include:
ƒ a Paprika field day with Zambia Association for Higher Value Crops
(ZAHVAC)
ƒ an outgrower scheme with Omnia based on maize production
ƒ an outgrower scheme with Castor company for castor beans
ƒ an outgrower scheme with Aminita for sunflower
ƒ a partnership with IDE for demonstrations of treadle pumps
ƒ linkages between vegetable production interest groups and urban vegetable
traders through the Export Board of Zambia
ƒ purchases by Oxfam and CLUSA of improved seed varieties
LFSP Mid-Term Evaluation ARD
77
8.5 Other NGOs
Other non-governmental organizations have been involved particularly in
providing services to the water section of LFSP. These include:
ƒ JICA involved in sinking boreholes
ƒ AFRICARE involved in sinking boreholes, with funding provided through
UNICEF and
Maintenance of boreholes and other water points is done through the V-WASHE,
through which LFSP provides training to selected individuals who become pump
minders and borehole caretakers.
8.6 SWOT Analysis of the Major Partnerships
8.6.1 MAFF
Strengths ƒ Conducive framework at national level
ƒ Similar technical focus
ƒ Need each other
Weaknesses ƒ MAFF not involved in formation of CBOs
ƒ No MOU
ƒ Based on good will and personalities
ƒ MAFF officers not completely oriented to
participatory methods
ƒ MAFF staff hampered by low incentive
structures and transport allowances
Opportunities ƒ MAFF has a long-term presence in the area,
can help sustain LFSP achievements after the
project completion date
ƒ MAFF has embraced participatory
approaches to extension
Threats ƒ LFSP could end without an adequate
transition phase
ƒ personnel could change, thereby changing
the dynamics of the relationship
8.6.2 Forestry
Strengths ƒ Compatible mandates
ƒ Complimentary resources
ƒ Compelling need for stronger cooperation
acknowledged by both sides
Weaknesses ƒ Forestry extension officers based in
Livingstone and don't get to travel to AMCs
as much
ƒ Same constraints as MAFF in terms of salary
incentives and transport
Opportunities ƒ Recent NRM proposal for Joint Forestry
Management
LFSP Mid-Term Evaluation ARD
78
Threats ƒ Expectations for the Joint Forestry proposal
are high. Confidence in the program might
be damaged if the Joint Forestry proposal
isn't enacted
Figure 63. To strengthen its institutional relationships, LFSP has recently conducted a series of two
partnership workshops, with a third one scheduled for June 2000.
LFSP Mid-Term Evaluation ARD
79
9.0 POLICY ISSUES
CARE/LFSP has been implemented against the background of far-reaching policy
reforms affecting virtually every public and private sector institution in Zambia.
These policy reforms have affected the agricultural, macro-economic, marketing, and
environment and natural resources sectors. The two policy arenas that have the
greatest relevance to LFSP are the agriculture and forestry sectors.
9.1 Agriculture Policy Context
The agriculture policies in Zambia are based on a Policy Framework to the year
2000. This framework was adopted by MAFF to formulate the Agriculture Sector
Investment Programme (ASIP) whose implementation started in 1996. This Policy
framework provides clear goals and strategies aimed at creating an environment
for increased private sector participation in agriculture. ASIP aims at facilitating
and supporting the development of sustainable and competitive agricultural sector
to ensure food security at national and household levels and to maximize the
sectors contribution to the national gross domestic product (GDP).
Government has adopted five basic objectives to transform the agriculture sector
from a highly controlled and regulated industry to one which is market oriented.
These objectives are:
ƒ Ensure national and regional food security through dependable annual
production of adequate supplies of food staffs
ƒ Generate income and employment
ƒ Ensure that the existing resource base is maintained and improved upon.
ƒ Contribute to sustainable industrial development.
ƒ Significantly expand the sectors contribution to the national balance of
payments.
Strategies developed for achieving the above objectives include the following:
ƒ market liberalization
ƒ crop diversification
ƒ provision of services to small-holders in outlying areas
ƒ development of the livestock sector
ƒ improving opportunities for outlying areas
ƒ making better use of available land
ƒ emphasising sustainable farming systems
ƒ improving the economic status of women
ƒ making better use of available water
ƒ helping farmers cope with natural disasters
9.2 Forestry Policy Context
The current forest policy, adopted in 1998, aims at increasing the country’s forest
cover and simultaneously meet the growing local needs for fuel wood, fodder,
LFSP Mid-Term Evaluation ARD
80
timber and other forest products (MENR, 1999). It encourages joint forest
management systems with the active involvement of local communities in the
protection, management and utilization of forest resources.
Key strategies developed to meet the forest management goals include the
following:
ƒ ensuring sustainable forest management
ƒ developing capacity of all stakeholders in sustainable forest resource
management and utilization.
ƒ promoting equitable participation by women, men and children in forestry
development and adopt an integrated approach through intra and inter￾sectoral co-ordination in forestry sector development.
9.3 Impact of Government Policy on LFSP
Current government policies have led to the liberalization of crop marketing,
removal of price controls on agricultural produce and inputs and the restructuring
of public expenditure on agriculture by removing subsidies. These policies have
given rise to crop diversification as farmers have had to adjust their production to
crops which have a comparative advantage in their respective areas. This has since
led to the expansion of traditional crops like sorghum, millet, cassava, sweet
potato etc. being grown in outlying areas (MAFF, 1999).
It is clear that the above government policies support and encourage partnerships
among different stakeholders including the participation of NGOs, private sector
and communities in development initiatives. To this effect therefore current
government policy provides a conducive environment in which LFSP can operate.
The market liberalization and removal of subsidies have impacted positively on
project implementation. The absence of input subsidies has, for instance, made it
easier to implement the loan schemes, in particular for seed provision.
Liberalization has also allowed the project to implement marketing strategies at
the local level (e.g., linking CBOs to trading agencies for input supply and buying
of farm products.)
Although there are no significant policy restraints on the project, there has been
some negative impact of MAFF policy on the project caused by the different
approaches to extension. LFSP found MAFF approach not effective and therefore
opted for a more community-based participatory approach. This difference in
approach has to some extent limited participation by MAFF staff at district and
camp levels in the activities of the project. This situation is not conducive for the
sustainability of interventions put in place under the project and runs contrary to
the strategic objective of both institutions to increasing private sector participation
in agricultural development.
The provisions of the current Forest Act presents an opportunity to CARE/LFSP
to embark on initiatives aimed at assisting communities to get involved in forest
LFSP Mid-Term Evaluation ARD
81
management and in sharing of benefits accruing from both commercial and non
commercial exploitation. LFSP is supporting the forest policy through its
activities in Chief Sekute’s area, which has included sensitization of the people on
the provisions and implications of the new forest policy and promoting
collaboration among different stakeholders. These activities have culminated in
the development of a project proposal on participatory forestry management in
Chief Sekute’s area. Through this proposal LFSP hopes to assist in developing
mechanisms to implement the policy.
9.4 Impact of LFSP on Policy
LFSP's overall strategy to improve food security is not to target every vulnerable
community in Zambia or even Southern Province, but to demonstrate new
approaches and new technologies in the hopes of making development policies
more effective and influencing the design of other investments. Accordingly, its
impact on policy is one of the more important but under appreciated components
of its operations. The project does not explicitly have goals for impacting policy,
nor any performance indicators on influencing development policy through the
dissemination of lessons learned.
Government views NGO participation in rural development as being important
and useful in helping to generate and adapt technologies that are well suited to the
conditions of the communities concerned. LFSP's overall focus on the underlying
causes of vulnerability and food insecurity is therefore very much supportive of
government policies of ensuring household food security and assisting farmers
affected by natural disasters. The specific strategic objectives developed under
LFSP (see section 4.0) are also very supportive of government policies.
One of the project's strategies that seems to be having, or may have, a major
influence on government policy is that of building capacity of community
institutions. Linked to this focus is the extension approach that LFSP has adopted.
MAFF for instance seems to appreciate the advantages of community based
participatory extension methods as opposed to the standard “training and visit”
system which it has been using (see also 10.1 on page 84). There is no doubt that
the success of the CBO approach in the delivery of extension service in the LFSP
areas is encouraging MAFF to modify its own extension system. Indeed the
MAFF Department of Field Services has recently developed a framework for a
pilot test of the participatory extension approach in 27 districts (MAFF, 2000)
with support from the World Bank, another die-hard advocate of the T&V system
which has only recently begun to see the light.
The current government policy on the provision of input loans and other
development support has also recently shifted from a focus on individuals to a
focus on CBOs. This can be seen in the MAFF infrastructure development support
programs under the Rural Investment Fund (RIF), and more recently the provision
of loans for livestock disease control program in Southern Province (Presidential
Fund). The Pilot Investment Fund is another micro-loan program being
LFSP Mid-Term Evaluation ARD
82
administered by the Environmental Support Program in the MENR which
administers to CBOs.
Although LFSP can't take the complete credit for the broader shift in government
to focus rural development interventions on small groups, the experiences of
LFSP have made an important contribution to the body of knowledge and
strategies for groups.
One of the strategic advantages of implementing development interventions
through a large international NGO like CARE are the corporate capabilities it
brings for influencing policy at the donor and international levels. CARE
presented the LFSP story at a 1997 Donor Consultation meeting in Rome,
attended by most major donors including the influential World Bank. CARE was
also invited to present LFSP at a follow up meeting in Ghana organized by the
FAO in 1998. They couldn't attend in person due to a last-minute logistical screw
up, but the paper was presented on their behalf. CARE staff also prepared a
technical paper for the ODI Agricultural Research and Extension Network
(AGREN) network paper series, which was published in 1997. CARE USA
presented a paper entitled “Participation within Participatory Technology
Development and Dissemination: Zambian Case Study” at the ODA/NRI
Workshop on Participatory Technology Development, Kenya, April 1997
(Drinkwater, 1997). The LFSP story has also been presented at various fora of
CARE International.
At the national level, LFSP has reached out a wide array of stakeholders to share
lessons learned and influence development policy. While the evaluation team was
visiting Livingstone, a 30-minute video of LFSP aired on the Zambia National
Broadcasting Corporation (ZNBC) network. This professionally produced video
provided an overview of the project's achievements and strategies. A glossy
booklet providing an overview of the project has also been recently published and
distributed to a wide range of stakeholders. LFSP has also hosted a long list of
visitors (see Table 7) and is commonly represented in national level meetings in
the agriculture and development sectors.
Year Institution Focus of visit
1998 WFP, USAID, REDSO/ESA preparation for food relief
Deputy Ministry MAFF follow up on food relief
USAID Southern African Office overview
SADC Seeds for Diversity project overview
CARE Canada food security strategies
CULP (CARE Zambia) overview
CIDA Vice President overview
CARE USA overview
British High Commission overview
Intermediate Technology Group, Chivi
Food Security Project
NRM
National Heritage and Conservation NRM
LFSP Mid-Term Evaluation ARD
83
Year Institution Focus of visit
Commission
Overseas Development Institute seed information systems
USAID Regional Office - Nairobi overview
1999 WFP follow up on food relief
CARE Malawi overview
PELUM Association M&E
Omnia partnering
Amanita partnering
Oxfam partnering
Export Board of Zambia market linkages
Action Aid Malawi - Smallholder Seed
Development Extension Project
CBOs
CARE Canada overview
1999 World Bank MAFF evaluation
IFAD/World Bank overview
DACO - Choma overview
SADC Ministers overview
Canadian High Commissioner overview
GTZ seed multiplication
2000 US Ambassador to Zambia overview
ZATAC dairy production
Provincial Planning Unit overview
Rotary Club, Livingstone activities in new AMCs
ADMADE partnering
Luapula Livelihood overview
SARNET tubers
Zamseed seed growers
Zamseed/MAFF/World Vision CBOs, seed multiplication
Table 7. Partial list of visitors to LFSP, 1998-2000.
Source: LFSP visitors book.
At a more local level, LFSP has played a role in shaping the development
strategies of partner institutions. MAFF has taken notice of the pros and cons of
LFSP's intervention strategies, and adopted the extension through CBOs approach
in some of its camps. LFSP is also represented on the District Development
Coordinating Committee (DDCC), and recently tried to strengthen its relationship
with other local institutions by sponsoring a series of partnership workshops.
Some CBO leaders in the LFSP areas have also been elected to the District
council because of their high profile in the community, an unexpected mechanism
that will surely accelerate the replication of LFSP's lessons learned within the
district.
LFSP Mid-Term Evaluation ARD
84
10.0 OTHER ISSUES
10.1 Extension
LFSP uses a CBO structure to deliver extension services for various development
interventions including improved farming practices, water harvesting and
sanitation, and natural resources management. These CBO structures have been
formed with full participation and consent of the people themselves. The CBO
model consists of a three strata structure (see Figure 11 on page 13). The smallest
unit is the Cell group, which consists of 4-7 households focused around a common
interest such as seed loans. The next level is the Village Management Committee
(VMC). Each cell group is represented on a VMC. An average of ten VMCs
federate to form the Area Management Committee (AMC) at the apex of the CBO
structure. Initially the cell groups were formed for the purpose of facilitating the
provision of seed loans.
Participation
LFSP's extension system is based on training needs determined primarily by the
beneficiaries, taking into account their own development priorities. Another
important characteristic of the LFSP extension approach is an emphasis on
participatory methods and farmer-to-farmer linkages. Under this approach, most
of the extension work is expected to be done by the rural people themselves, with
the assistance of local facilitators who are selected by their fellow community
members and trained by the project. Extension services are further enhanced
through farmer-to-farmer learning within cell groups.
LFSP Extension Officers
LFSP has about 12 extension officers for the various project sections. They are the
front-line staff linking the project and rural communities. One of the main roles of
extension officers is to facilitate community organization and capacity building.
Extension officers also play a major role in providing the initial training to the
community facilitators. Although all extension officers have a technical
specialization and are able to backstop the communities on technical issues, they
are also expected to have a broad outlook in line with the project’s holistic
livelihood approach to development. LFSP extension officers, unlike MAFF camp
extension officers, do not reside within the communities they serve, but are based
at the project offices either in Livingstone or Kalomo. The frequency of contact
with the community is therefore relatively low.
Community Facilitators
Facilitators are an integral part of the CBO extension system under LFSP. The
facilitators are selected by their peers within communities at the VMC or AMC
levels and form an important link between the people in the communities and the
LFSP extension officers. In general, facilitators are energetic and knowledgeable
young people who play a big role in capacity building for community participation
in providing extension services to their own communities. There is at least one
facilitator in each AMC. Facilitators are not paid, although in a few instances they
have been nominally compensated by the VMCs they serve.
LFSP Mid-Term Evaluation ARD
85
Farmer-to-farmer learning is facilitated through field days, which are organized by
the communities with the help of the project. The project encourages farmer
exchange visits between AMCs in order to promote farmer to farmer learning
through on-farm demonstrations.
Technology Demonstrations
The LFSP extension system includes the use of demonstrations on selected farms.
The purpose of these demonstrations is twofold, one being to test the technology
for adaptability to the specific conditions of the project, area while the other is to
show the desirable attributes of the new technology and hence the benefits it can
bring to the people.
On-farm demonstrations of improved technologies have been shown to be an
effective way of promoting wider adoption of such technologies. There are a
number of technologies related to sustainable farming practices, such as the use of
green manure and agroforestry tree species for soil improvement, that have only
been tested on a small scale through demonstrations in the project area (see Annex
V). It will be necessary to expand the demonstration program to additional area if
chances for their wider adoption by farmers are to be increased.
The strengths and weaknesses of the LFSP extension approach are summarized in
Table 8 below.
Strengths Weaknesses
1. Low running costs
2. Demand driven as the approach is
based on training needs identified
by the people
3. Builds local capacity of the
communities to participate in
training among themselves
4. Encourages learning by farmers
from each other
5. Groups are self-sustaining as they
are based on identified interests
6. Enables reaching more people more
quickly and with limited resources,
and therefore tends to be more cost￾effective
1. Inadequate supervision of
implemented activities as a result
of reduced contact between project
staff and the people in the
communities
2. High start-up costs, mainly arising
from PRA exercises and CBO
training
3. CBOs not based on existing
structures and may therefore lead
to conflicts with existing
institutions which could affect
effectiveness and sustainability
4. LFSP extension staff may not have
all the technical know-how
requested by beneficiaries
Table 8. Strengths and weaknesses of the LFSP CBO approach for extension
MAFF Extension Approach
The official approach to agriculture extension in Zambia has been the Training
and Visit (T&V) system. This system was introduced through the World Bank
supported Zambia Agricultural Research and Extension Project (ZAREP). Under
the T&V system the camp is divided into four zones with two farmer groups
formed in each zone. There are therefore eight farmer groups that a camp officer
LFSP Mid-Term Evaluation ARD
86
has to work with. This was a move away from the contact farmer concept to group
teaching. The T&V system is characterized by extension officers following a strict
and rigid program of visitation with the farmer groups The camp officer visits
each farmer group once fortnightly and trains them using technical messages
provided by the district subject matter specialist or section head. The messages are
supposed to be based on questions raised by farmers from the problems that they
experience. The system therefore has a feedback mechanism, which allows for
some measure of farmer participation in the training process. The T&V system
also includes monthly training of camp extension officers as they receive technical
messages from subject matter specialists. This also provides an opportunity to the
camp officers to be updated on new technologies.
However experience with the T&V system has shown that it is expensive to
implement, is ‘top down,’ and stifles self-initiative among field staff and farmers.
The system also promotes the belief that extension officers are the teachers while
farmers are mere recipients of knowledge and skills, thus making the approach
more technology focused rather than people focused.
There has been a general realization that the T&V extension system, like other
extension systems tried in the past, has had little positive impact in terms of
adoption of the technologies by the farmers. This has further led to the realization
that there is need to seek more incorporating systems that promote increased
farmer participation at every stage of decision making process. The result of this
realization has been a proliferation of approaches that sought to modify the T&V
system or to adopt more participatory approaches that would employ interactive
dialogue and involve rural people and institutions in development planning and
implementation. A variety of participatory approaches to extension have been
adopted or tried in many parts of the country in an effort to accelerate adoption of
improved technologies. Individual donors, under specific projects have driven
many of these initiatives.
MAFF currently seems to be moving in the direction of promoting more
participatory approaches to extension, however there is no official
acknowledgement on whether the T&V system is giving way to Participatory
Extension Approaches. However the MAFF Department of Field Services has
recently developed a framework for the implementation of Participatory Extension
Approach (PEA) in 27 pilot districts with support from the World Bank (MAFF,
2000).
Cost Effectiveness
The LFSP extension approach is expected to be more cost-effective compared to
the T&V extension system. The evaluation team is, however, not in a position to
compare the actual costs of delivering extension service between the two systems
due to lack of information. However an attempt could be made to compare the
effectiveness of the two systems in terms of number of farmers reached by one
extension officer.
LFSP Mid-Term Evaluation ARD
87
LFSP estimates that by working through CBOs one extension worker can work
with approximately 1000 farmers (LFSP, 1999a). In comparison, an average a
MAFF camp officer under the T&V system only works with 300-500 farmers.
Furthermore, the T&V system has been found to be expensive to implement due
to a regular and rigid visitation schedule, which may not always be needed, and
relies on the availability of transport for the camp officer. Although the camp
officer lives within the communities, lack of logistical support and low
compensation levels can affect the effectiveness of the T&V approach, as some
officers must spend a lot of time doing farming to support their own families.
10.2 Seed Scheme
Loss of local seed stocks after repeated years of drought and low production was
one of the main factors identified in the formative PRA exercises in 1995. Hence
the seed multiplication scheme was a natural choice for the first and central
intervention under LFSP.
10.2.1 Seed Selection
The first step taken in the implementation of the seed scheme was the
introduction of drought tolerant and early maturing varieties of a number of
crops, including maize, sorghum, cowpea, groundnuts and pearl millet. See
Annex VI for a complete list of the improved seed varieties introduced and
their specific characteristics. LFSP deliberately concentrated on open￾pollinated varieties and avoided hybrids in its seed multiplication scheme. The
wide range of introduced seed varieties had the effect of diversifying crop
mix, one of the strategies used to achieve sustainable farming in the project
area.
10.2.2 Impact on Food Production and Availability
The main agronomic attributes of the various improved crop varieties
introduced in the project under LFSP are early maturity and high yielding
compared to the local varieties. The maturity range for all crop varieties that
are included in the seed scheme is 90-110 days from sowing to harvest. The
estimated yield potentials of the improved crop varieties introduced into the
project are 4.0 tonnes/hectare for maize, 2.0 t/ha for sorghum, and 2.0 t/ha for
cowpea. However the average yields among LFSP farmers in the project area
during the 1995-99 period were still relatively low, at 1.04 t/ha for maize, 0.6
t/ha for sorghum and 0.25 t/ha for cowpea.
There has been an increase in the period during which households have food
in the project area since the seed scheme under LFSP began. For instance,
34% of the households in the project area were by 1999 maize secure over a
period of 9 months compared to only 5% having food for less than 6 months
before 1995 (LFSP, 2000). It has also been reported that most of the food
available to households for consumption is derived from own production
(Ndiyoi et al., 1999). It can be assumed that the extended periods of food
availability at the household level have resulted in large part from the
increased crop yields.
LFSP Mid-Term Evaluation ARD
88
The early maturing crop varieties of maize and cowpea, which may be ready
for harvest around January, are reported to have become an important source
of food during the hunger months of January through to March (LFSP, 1999e).
It is expected that much of the produce from the early maturing varieties of
maize, sorghum and cowpea is consumed during the hunger period leaving
little to put in the granaries. This allows the harvest period to focus on plots
planted with local varieties and hybrids, to be stored and consumed during the
rest of the year. There are indications that more area is being planted to these
improved early maturing varieties of maize, and are in some cases, even
providing the bulk of the harvest, thereby becoming the main source of food
for most part of the year.
10.2.3 Seed Quality Problems
Declining seed quality has been identified by LFSP staff as a problem that
could be contributing to a decline in the yields of certain crops (LFSP, 1999e).
An analysis of the CSM data by the evaluation team also detected this trend.
One of causes of poor seed quality is that farmers have a tendency to pay back
seed loans with low quality seed. This could be caused by delays in collecting
seed from farmers, or deliberate efforts to repaying with inferior quality seed.
The same assessment of the seed multiplication scheme noted that higher
quality seed was used in farmer to farmer bartering and exchanges.
Figure 64. Seed repayment requires some quality control measures
Poor seed germination has also been observed from seed kept in seed banks
within VMCs and AMCs. This could be associated with storage problems
resulting in insect pest damage (weevils for maize and sorghum and brucchid
for cowpea) and inappropriate seed moisture and temperature conditions.
Other associated problems could be varietal degeneration, especially in open
LFSP Mid-Term Evaluation ARD
89
pollinated crops such as maize, as seed is recycled from one generation to the
other.
As CBOs became more independent in managing the multiplication and
distribution of seed, the project anticipated more problems related to seed
quality. To reduce such problems, the project sought the assistance of the Seed
Control and Certification Institute to conduct inspections of the seed crops
once a year, mainly at the time of harvest, and get involved in providing
training to farmers in aspects of seed production, handling and storage.
LFSP may have to try new approaches to address the issues of long term seed
quality. One option is to select a few farmers to produce seed under more
tightly controlled conditions for several VMCs or AMCs. These farmers will
require to be linked to the official seed certification institutions to ensure that
the necessary seed quality factors are taken into account. The other important
linkage for the seed multiplication arrangements at the community level is the
source of foundation seed. This may either be breeders under public research
institutions or private seed companies.
10.2.4 Sustainability of the seed scheme
Indications from LFSP reports and the internal assessment of the seed scheme
are that the supply seed for different crop varieties has not been able to meet
the demand for seed in the project area (LFSP, 1999e). If anything, the
demand for seed is on the increase even in areas where the seed scheme has
been operational for a number of years. One would expect more demand for
legume crops which also improve soil fertility, especially those with higher
economic value such as groundnuts.
The initial objective for the introduction of early maturing drought tolerant
crop varieties in the project area was to mitigate drought and alleviate hunger
among the affected communities. There is little doubt that the crop varieties
introduced and which are included in the seed multiplication scheme have
been appreciated by the farmers, evident by their wide spread adoption. The
broad demand for seed will play an important role towards the sustainability of
the seed scheme even after LFSP comes to an end.
The other factor which will help ensure the sustainability of the seed
multiplication scheme is the realization of economic benefits from increased
yields from improved seed. Increased yields will enable people have surplus
produce, which they can sell to raise money if market linkages or informal
trading networks are in place (Kalonge & Pongolani, 1997). However there
may be little incentive for increasing production and realizing the potential
economic benefit if marketing of surplus is not improved. Currently there are
limited outlets for agricultural products in the project area. The consequence
of not developing markets and market linkages may be reduction in the area of
cultivation for the drought tolerant, early maturing and high yielding varieties.
LFSP Mid-Term Evaluation ARD
90
This may have a negative impact on crop diversification efforts and eventually
on the sustainability of the seed scheme.
Sustainability of the seed multiplication scheme in the project area may
benefit from varying it from its present form. It could be desirable to allow it
to evolve from being based on loans and community seed banks to individual
seed enterprises. The project could therefore pay deliberate attention to
developing capacity of selected farmers with potential to become seed
producers and dealers within the communities. Other efforts that may be
considered in order to sustain the seed scheme is expanding markets beyond
the project areas where surplus seed stocks may be sold.
10.2.5 Community Seed Banks
The seed bank concept was developed
as an integral part of the LFSP seed
scheme and its purpose has been
appreciated by most of the farmers in
the project area. The community seed
banks, which are located at VMCs and
AMCs, are used for the storage of seed
received both from LFSP and the
farmers through loan repayments. Seed
distribution to farmers for planting is
also distributed from the same seed
banks. One of the major benefits of
community based seed banks is that it
facilitates early distribution of seed to
farmers and guarantees seed
availability at planting time. Prior to
LFSP and the introduction of seed
banks, seed stored in homes by
individual households used to be
vulnerable to be consumed as food
during hunger periods.
Figure 65. Local seed banks ensure that
seed is available for planting
The role of community seed banks is also being extended to store seed from
selected seed growers within the community.
10.2.6 Diffusion of seed
Milimo and Tripp (1999) studied seed use outside LFSP areas, and found that
there is a significant amount of diffusion of seed and information from LFSP
farmers to other areas. Within 10 km of LFSP villages, 45% of sampled
farmers got their sorghum seed from LFSP villages. At distances greater than
20 km, 21% got their seed from LFSP farmers. While the magnitude of this
trade is difficult to estimate, its effects should be seen on the food availability
of nearby areas (which is currently not being monitored). These additional
farmers can be considered to be indirect beneficiaries of LFSP.
LFSP Mid-Term Evaluation ARD
91
Distance from LFSP
village
Number of
sampled
farmers
LFSP village
as source of
seed
LFSP village
as source of
information
First saw
variety in
LFSP village
Less than 10 km 53 45% 42% 62%
More than 20 km 34 21% 21% 29%
Table 9. LFSP villages as sources of sorghum seed and information.
Source: Milimo (1999).
Although the project has not been able to capture the multiplier effects of
secondary seed trade very precisely, one way the informal trade of seed can be
better monitored is by including seed sale and trade in the CSM ledgers. This
additional column would also help the project detect trends in selling strategies
(e.g., how long farmers wait before selling), and the impact of increased
production on rural household. The additional column in the CSM would not be
too onerous to collect, and indeed some VMC secretaries have already added
columns for seed sale and trade to the CSM on their own initiative (Lyons,
1998a).
10.3 Monitoring and Research Issues
10.3.1 Need for Production Data
In the evaluation team’s search for production data relating to: (1)
dissemination of drought tolerant varieties; (2) soil conservation technologies
evaluated in demonstration plots in farmers’ fields; and (3) for vegetable
crops, no quantitative project data regarding production and area cultivated,
was discovered at the LFSP office in Livingstone. Without quantitative
production data, it is difficult to evaluate the technologies. Some area farmers
may be able to visit the demonstration plots (vegetable gardens, etc) but
certainly further away farmers could benefit from such quantitative data. The
project also requires production data including the reporting of results to
audiences far and near to the project area. The project has made plans to
conduct crop-cuts on a pilot basis, which is a good move towards getting
better information.
10.3.2 Use of estimators
The project has done a remarkable job of studying the different units used by
farmers for their own calculations of production data, including traditional
area measurements such as ‘folo’ as the area a team of oxen can plow in a
good morning’s work. The ‘folo’ is roughly equivalent to 0.2 hectares. See
(Milimo et al., 1997) for a list of the local units for weights, sizes, etc.
The project makes extensive and routine use of estimators when measuring
reporting project impact or production trends. Table 10 and Table 11 below
list the main estimates used.
LFSP Mid-Term Evaluation ARD
92
Crop Seed
applied
(kg/ha)
Seed per
beneficiary
Area per
beneficiary
Crop yield
(kg/ha)
Production
per beneficiary
(kg)
Maize 20 10 0.50 2,300 1,150
Sorghum 8 4 0.50 1,400 700
Pearl millet 8 2 0.25 1,000 250
Groundnut 80 10 0.13 1,200 150
Cowpea 15 2 0.13 1,500 200
Bambara nut 80 10 0.13 1,200 150
Velvet bean 50 10 0.20 3,000 600
Table 10. Estimations used by LFSP for area planted and production
Use of estimators is often necessary when actual figures are either unknowable
or unfeasible to obtain. However a few cautions are to be noted about using
approximations for analyses:
ƒ The approximation itself should be based on real data, with outliers
and suspicious data thrown out.
ƒ Estimators based on real data always have an associated confidence
interval, in other words a range of probable values instead of just a
single number. Consequently any calculations based on that indicator
should also be based on ranges.
ƒ Estimators are often specific to a given year or region. For example,
according to the CSM database, maize yield for improved varieties was
indeed 2,300 kg/ha in 1996, but it came down to almost half in the
following two years.
ƒ Errors in estimators tend to get magnified the more they are propagated
through equations.
ƒ Performance measures based on estimators should be clearly noted as
such.
ƒ Whenever possible raw data used for analyses should be saved so that
it can be reviewed later.
Number of beneficiaries per VMC: Estimates versus actual
As much as possible, LFSP tries to get an up to date count of the number of
members of seed groups to calculate the number of beneficiaries. However it
is often necessary to use estimates when real data is not available (see Table
11).
Variable Estimation
Number of households per VMC 15
Number of beneficiaries per household 3
Number of beneficiaries per VMC 45
Number of inhabitants per household 6
Table 11. Estimations used by LFSP for beneficiaries
LFSP Mid-Term Evaluation ARD
93
To illustrate the dangers of using averages as estimates, Table 12 and Figure
66 below show actual data for the number of beneficiaries per VMC presented
as a table and histogram.
Center VMCs Beneficiaries Beneficiaries
per VMC
7A 2 204 102.0
Bbilili 14 660 47.1
Busanga 2 134 67.0
Chabalanda 2 72 36.0
Dundumwezi 5 496 99.2
Katapazi 18 636 35.3
Libala 7 120 17.1
Makunka 37 1,583 42.8
Mandandi 18 528 29.3
Mandia 13 504 38.8
Milangu 5 726 145.2
Mukuni 8 1,456 182.0
Musokotwane 10 348 34.8
Muzumbwe 5 144 28.8
Mweemba 10 820 82.0
Sekute 30 700 23.3
Shindu 7 216 30.9
Siakasipa 6 646 107.7
Siamasimbi 8 222 27.8
Siandasya 7 264 37.7
Sihumbwa 15 754 50.3
Sinde 13 598 46.0
Table 12. Number of beneficiaries per VMC.
Source: (LFSP, 1998a)
LFSP Mid-Term Evaluation ARD
94
Beneficiaries per VMC Histogram
0
1
2
3
4
0-5
10-15
20-25
30-35
40-45
50-55
60-65
70-75
80-85
90-95
100-105
110-115
120-125
130-135
140-145
150-155
160-165
170-175
180-185
Beneficiaries per VMC
Frequency
Average = 49
Mode = 41
Average minus two outliers = 42
Figure 66. Beneficiaries per VMC histogram
Source: (LFSP, 1998a)
As can be seen in the histogram, the number of beneficiaries per VMC does
not follow predictable distribution like a bell-curve. Furthermore, there are a
couple of outliers that throw the average way off. Thus depending on how
accurate you want to be, the average value of 49 beneficiaries per VMC would
be a good estimate for a few of the VMCs, but very far off for many others.
When the average is used as an estimate for the number of beneficiaries for
several VMCs, the potential for significant error increases, and without the use
of confidence intervals there's no way to tell how accurate the performance
indicator is. Often the median is a better measure of the central tendency of a
dataset, because it is less sensitive to outliers.
Table 13 below gives another example of the trouble of using estimates.
Production yields vary enormously between years and agroecological zones.
The use of a single estimator for production yield is simply not very reliable.
Food Production
Trends Survey
Yields (kg/ha)
Agro-Zone
1995/6 1996/7
LFSP yield
estimate
(kg/ha)
Avg.
difference
real minus
estimate
Average
% real of
estimate
Maize
Escarpment 1,849 1,082 2,300 -834 64
Kalahari 1,071 825 2,300 -1,352 41
Plateau 1,831 900 2,300 -934 59
Sorghum
Escarpment 839 95 1,400 -933 33
Kalahari 926 219 1,400 -827 41
Plateau 1,082 674 1,400 -522 63
LFSP Mid-Term Evaluation ARD
95
Food Production
Trends Survey
Yields (kg/ha)
Agro-Zone
1995/6 1996/7
LFSP yield
estimate
(kg/ha)
Avg.
difference
real minus
estimate
Average
% real of
estimate
Pearl millet
Escarpment 372 235 1,000 -697 30
Kalahari 1,091 716 1,000 -97 90
Plateau 564 556 1,000 -440 56
Groundnut
Escarpment 121 249 1,200 -1,015 15
Kalahari 877 892 1,200 -316 74
Plateau 564 677 1,200 -579 52
Cowpea
Escarpment 471 693 1,500 -918 39
Kalahari 225 162 1,500 -1,306 13
Plateau 23 124 1,500 -1,426 5
Table 13. Real and estimate production data.
Source: Food Production Trends Database.
Summary and Recommendations
Sometimes the use of estimates is required because no real data exists.
However LFSP often has real data available which could be used to produce a
more accurate picture of project impact. The use of simple statistics such as
95% confidence intervals would help readers interpret the validity of
performance measures. At a minimum, making greater use of histograms to
show the distribution of data can help LFSP staff and readers interpret data
and look for outliers. Histograms such as the one in Figure 66 above can easily
be created in Excel using the Analysis ToolPak add-in.
10.3.3 Use of case studies
All sections of LFSP involved in research make extensive use of the case
study method to assess how interventions are performing. A few notes
concerning this research approach are worth mentioning:
ƒ The case study is an excellent research method to study process.
However it is much less suited for assessing impact.
ƒ Case studies can supplement systematic monitoring and evaluation
methods, but can't replace them.
ƒ The findings of case studies can be made more valid by using random
selection in the sampling of objects to observe.
ƒ Comparative case studies (e.g., weak and strong AMCs) provide
multiple examples of a process at work, thereby increasing the validity
of patterns and cause-effect relationships detected.
ƒ The use of case study in research does not mean that you can run away
from collecting quantitative data. A good case study will use
quantitative information to verify qualitative statements and support
conclusions drawn.
LFSP Mid-Term Evaluation ARD
96
10.3.4 Sampling
Sampling is a standard practice in research because it often isn't possible (or
necessary) to measure each and every household, VMC, field, etc. The general
objective in sampling is to get a subset of the population which provides a
balanced representation of the total population. That means that the subset
must be selected in such a way that it is a fair reflection of the population.
Random selection is often the preferred method because it ensures that
whatever variation exists it will at least average out in a predictable way.
How large should our sample be? is a common question asked by research
teams. Although you can use statistical procedures to tell whether a sample
size was adequate after the data has been collected, there are no hard and fast
rules about the perfect sampling size. There are however a few guidelines:
ƒ Before you can know the appropriate sample size, you have to define your
desired confidence limits. For example if you say that you want the food
availability estimates to be 95% accurate plus or minus one month, and
you know the variance of sample data, then you can estimate the number
of required samples.
ƒ The confidence one can have in findings of a study is related sample size
through well-known statistical formulas. However when statistical
methods (e.g., confidence interval) are not used in analysis (as in the case
of LFSP), then you can only make a qualitative assessment of the validity
of results. In this case common sense and good judgement plays as much
as a role as sample size.
ƒ As a ball park figure, 30 is the minimum number of randomly selected
observations that should be made for social science research3
. One fifth of
all objects, which is the sampling rate used by LFSP for some of its
studies, is also a good rule of thumb to use. However larger sample sizes
will provide more precise estimates.
ƒ When there is a lot of variability in the objects being studied (e.g.,
household production), then its better to have a larger sample size. When
there is less variation (e.g., household coping strategies), then the sample
size can be smaller.
ƒ When objects are selected in such a way as to ensure that they represent a
balanced selection (e.g., through stratification based on agroecological
zone), then the sample size can be a bit smaller.
ƒ If objects are selected in a non-random way (e.g., opportunistically or
based on convenience), then sample size should be higher (although there
will still be bias).
ƒ When you want to track change over time, to look for the effect of a
treatment, it is important that the same objects be measured.
Sampling under actual field conditions is always a challenge. LFSP uses
sampling in many studies, and has a better sampling program than many
projects because they have a good sampling frame (e.g., a list of all AMCs,
 
3
 in order to know that two standard deviations from the mean represents a 95% confidence interval
LFSP Mid-Term Evaluation ARD
97
VMCs) and have identified important variables for stratification (e.g.,
agroecological zone). LFSP has also used random selection of VMCs in
various studies.
The sampling process should always be well documented in research reports.
Bias in sampling is inevitable in most real-world situations, and should not be
viewed as a flaw. However potential bias in the sample should always be
acknowledged and addressed in the methodology section of reports.
Sampling and Measurement Problems in the Food Production Trends Survey
The Food Production Trends was based on a sample of 220 households
randomly selected (every 5th household of every 5th VMC) in 1996. This was a
promising baseline assessment that was to be updated once a year to track
impact of the project interventions on food production, crop mix, and food
availability. However there have been a couple of sampling and measurement
practices which jeopardize the validity of the Food Production Trends dataset.
Production data for the 1995/96 and 1996/97 planting seasons were both
collected in 1996/97. The problem with collecting two years worth of
production data at once is that it relies on recall of the previous year. Although
LFSP farmers might have photographic memories of the past planting seasons,
many studies have shown the in general recall data is not very trustworthy,
particularly if its more than a couple of months old.
A second questionable practice came when the same households were visited
the following two years. If the participating farmer was not at home, or had
moved or died, the enumerator went to the next door neighbor and collected
their data. This practice might not be so significant for analyses of variables
that don't differ significantly among neighbors (e.g., family size, language
spoken), however it greatly reduced the validity of any trend analysis of
variables which are likely to vary even between neighbors, such as area under
cultivation or production per crop. And tracking change is the most interesting
use for the dataset. Currently there are 138 households (over 1/2 the sample) in
the database which have changed between 1998 and 2000. Furthermore, there
isn't a column to indicate which households have substituted values.
However all is not lost. If the M&E Unit can find the original 220 households
that were visited in 1996/97 (or as many as possible), disregard the 95/96 data
as not trustworthy because it was based on memory, and ask the same set of
questions for the 1999/2000 season, then there would still be a good
before/after picture of the impact of LFSP. The intervening years might be
missing for many of the households, but at least changes in food availability
and production strategies over the four-year period would be valid. This would
be a helpful analysis for the final project evaluation.
The M&E Unit is also soon to be confronted with another sampling issue, as it
begins to enter and analyze records from the CSM. Here again the general
principles apply: when tracking changes over time, when at all possible the
LFSP Mid-Term Evaluation ARD
98
same households should be measured. Some of the interactive graphs built in
to the CSM database are designed to only include data from households that
have complete records for all of the years desired (e.g., see Figure 58 on page
70). Thus when selecting which CSM records should be entered into the
database, the M&E Unit should use two guiding criteria for prioritization:
1. CSM books that have multiple years of data for the same households
2. CSM books that represent all three agroecological zones
When analyzing results from the CSM database, the M&E team should also be
on the lookout for small sample sizes and outliers. For example, the average
value of all household assets for the escarpment zone for 1997 was several
times higher than any of the other zones. Upon closer inspection, it was found
that there were two households for that year which had large herds of cattle,
and this was throwing the average off because the sample size was only about
20. Fortunately MS Access queries can be designed to eliminate these types of
outliers for more valid results.
10.3.5 Information Technology
One of LFSP's principle strategies for improving food security is the
development and demonstration of improved technologies, including soil
conservation and fertility, new crop varieties, water harvesting structures,
small scale industries, post-harvest processing, and NRM. After identifying
promising develops these technologies through demonstrations and pilot
programs.
The effectiveness and environmental and labor requirements of new
technologies are identified through monitoring of the demonstrations. Both
beneficiaries and project staff need to see and understand the results. LFSP's
need for information on trials also extends beyond the project, for example its
institutional partners. Consequently there is a widespread need for
systematically collected and properly analyzed data on trials.
LFSP current capacity in IT is not up to the task of such an important and
large scale undertaking. Although all program staff are computer literate and
use word processing and spreadsheet software in their daily activities, the
project does not have the capacity to set up the types of multi-layered
monitoring databases required to adequately process the amount of data that
should be coming in from technology trials. The lack of real production data,
under-utilization of the CSM and Food Production Trends surveys, lack of
hard data on adoption rates, and over reliance on isolated case studies are just
a few examples of where limited IT capacity has constrained the ability of the
project to synthesize results and communicate lessons learned effectively.
If not strengthened, the inadequate capacity to manage quantitative datasets
may affect the ability of LFSP to build lasting relationships with partner
organizations, particularly those with a technical focus such as MAFF. The
evaluation team recommends that LFSP increases its IT capacity, either by
LFSP Mid-Term Evaluation ARD
99
developing internal resources or forming a tech service contract with an
appropriate firm.
10.3.6 Documentation and Dissemination
LSFP has been prolifically documented, largely by its own highly capable
staff. However a 1998 report concluded that LFSP's action research
documentation was its "best kept secret" (Lyons, 1998b). Turner (2000) also
identifies a need for organizing and improving dissemination of research
summaries and other project documents. LFSP acknowledges these sentiments
and has responded by recently creating a position for a documentalist. The
documentalist has a multifaceted job description which is part librarian, part
public relations, part journalist, and part editor.
One mechanism not currently being capitalized upon is electronic
dissemination of documents. Virtually all staff are keyboard literate, and use
Word Processing and spreadsheets in preparation of their reports. Hence it
would be a fairly simple matter to compile these electronic files in a common
location, for eventual distribution on a project CD ROM (which can contain
100's of reports and be reproduced for less than $2 per CD) or web site.
Although the LSFP may not have a web site presently, or even a compelling
need to distribute documents electronically, the odds are very high that the
project or CARE/Zambia will one day wish to make its many excellent
quarterly reports, PRA exercises, proposals, training summaries, etc., available
to external partners for minimum reproduction cost. LFSP can prepare for this
eventuality by beginning to stockpile documents today.
The quality of research documentation coming out of LFSP is fairly high,
however there is still room for improvement in certain areas. Research
methodology, including sampling schemes and measurement methods, is often
described only briefly or omitted completely. Sometimes the calculations used
to prepare a graph are not obvious and not documented (e.g., the definition for
a direct and indirect beneficiaries). Maps showing locations of project
activities would help orient readers to the research, and are now within the
capacity of the project to produce. Although most datasets presented in reports
are analyzed with fairly simple statistics, such as averages or graphs, some
data could be analyzed with slightly more rigorous techniques, allowing a
confidence interval to generated. The M&E unit already uses an excellent
statistics package, SPSS, to analyze data, so doing more robust analyses is
well within the capabilities of the project.
10.4 Scaling up Technology
The two most important strategies being used by LFSP to improve food security
are 1) building the capacity of CBOs, and 2) developing improved technologies.
The project is well on the way to achieving the first goal of building the capacity
of CBOs. The project has invested heavily in CBO training, and ten AMCs have
already been graduated. Although additional work is needed, there are many
LFSP Mid-Term Evaluation ARD
100
indications that the AMCs supported by the project will continue to evolve into
viable activity-oriented group enterprises.
The project has also made many accomplishments in developing improved
technologies in water harvesting and agriculture. However it has yet to implement
a strategy to achieve the next important hurdle: scaling up the adoption
technologies to achieve meaningful impact. To date, almost all of the water
harvesting, NRM, and soil fertility technologies have been limited to
demonstrations on a fairly small scale. This was a very reasonable approach as
many of the technologies had been demonstrated elsewhere but never been tested
in the project area. However now the project has a substantial body of technical
knowledge drawn from case studies, and is in a position to start replicating these
experiences in other areas.
The need for a scaling up strategy can not be overstated. There is a tremendous
need throughout the project area and in adjacent districts for improved methods
for smallholder farmers to improve food production and better manage natural
resources including water. The project's accomplishments in demonstrating many
of these technologies is laudable, but demonstrations in themselves have had
marginal impact on the overall food security situation. If the only output of the
many technology trials was a Lessons Learned document which did not translate
into wider replication and adoption, it would not be a good return on USAID's
$3.6 million investment in LFSP.
The project has made a good start in devising a strategy to replicate the lessons
learned from technology trials, particularly for NRM. It has documented the major
technologies tested, and identified the climatic, economic, social, topographic,
educational, and maintenance requirements of each of the NRM technologies
(Mwiinga, 2000). They have also identified factors which contribute to the
adoption of NRM methods by small scale farmers (see box below). They are also
in the process of publishing a resource handbook for many of the techniques used
in community capacity building, soil fertility, and income generation.
Factors Contributing to the Adoption of NRM Technologies
ƒ Benefits can be easily seen or demonstrated, (e.g.,
increased water, crop yields, income)
ƒ Investment costs with the reach of farmers
ƒ Availability of local materials or expertise
ƒ High levels of awareness and knowledge on the benefits
and management of the technology
ƒ Other farmers have experience with the technology
Source: (LFSP, 1999b)
Although the technical side of technology adoption is becoming increasingly
clearer through action research activities, particularly for NRM, more work is
needed in documenting the technical aspects of income generation activities, post￾harvest management, livestock management, and to some extent water harvesting
structures. Additional work is also needed in all sections in developing delivery
LFSP Mid-Term Evaluation ARD
101
mechanisms and an institutional network for the expanded application of new
technologies.
LFSP does not have the resources to expand the scope of its services much further
than its current geography, nor should it. Donor funded projects rarely have the
longevity or resource base to be responsible for the direct provision of services on
a long-term or widespread basis. The better strategy for scaling up technology
demonstrations is through partnerships with more permanent institutions, namely
government agencies and the private sector.
While LFSP acknowledges the importance of forming partnerships with
government agencies and private sector to achieve sustainability, these are also the
two lines of action where its has encountered the least amount of success.
Working with institutions and the private sector can be at least as tedious and slow
as working with rural communities, if not more so, and in some cases the
obstacles seem just as entrenched as the changing weather patters. Working with
institutions and private sector markets also requires different skill packages and
intervention strategies, which may not be very well represented in the project's
staff mix and overall orientation.
The evaluation team recommends that LFSP use the remaining period of USAID
financing to give proper attention to developing a long-term strategy for
developing delivery mechanisms for replicating technology trials through
partnerships with government and private sector institutions. Investments such as
the hiring of a marketing director, reorienting the focus of the SEAD section to
strengthening market linkages, and supporting sensitization programs in
participatory extension methods for MAFF staff are excellent initiatives in this
orientation. However these budding efforts still seem dwarfed by the dominant
focus on community capacity building and new technology development, and
caution must be advised that the inevitable phase out of the project not be
forgotten.
10.5 Life After LFSP
LFSP is approaching the last year of its current financing agreement with USAID,
and entering a period of intensified discussion as to what will become of the
project's achievements and capabilities. The overall objectives for LFSP for the
remaining funding period and extending past the project completion date include:
1. maintain the achievements already made in improved food security and
increased rural income
2. expand the number of beneficiaries in the existing service area
3. replicate the LFSP strategy and lessons learned in new areas
This section tries to explore some of the main issues involved in the post-LFSP
transition phase, and present some possible scenarios for the project's legacy in
Kazungula, Livingstone, and Kalomo districts. Unfortunately there is no crystal
ball that will provide a definitive answer on which direction will yield the best
LFSP Mid-Term Evaluation ARD
102
results, but at least the major stakeholders can begin to use analyses like this one
to increase the level of dialog on post-LFSP strategies.
Project Services
A discussion on post-LFSP needs to begin with a summary of the main roles and
services of the project, as these are the particular goods and services whose future
is under question. These are summarized in the table below:
Major Services Provided by LFSP
Support for CBO formation
ƒ PRA exercises
ƒ training in CBO structure and leadership roles
ƒ training in monitoring
ƒ assistance with problem-solving
ƒ CBO assessment
Support for improved farming systems
ƒ identification and sourcing of improved seed varieties
ƒ supplies of new seed
ƒ monitoring of seed multiplication activities
ƒ agricultural extension services
ƒ materials and training for demonstrations of improved
farming practices
ƒ on-site inspections and monitoring of trials
Support for water and sanitation projects
ƒ topical appraisals
ƒ technical advise, training, and materials for the construction
and maintenance of water harvesting structures
ƒ watershed management advice and planning
ƒ sourcing of spare parts
Support for income generation activities
ƒ training in business management
ƒ training in specific IGA skills (e.g., beer brewing, dry season
vegetable production, food processing)
ƒ training, materials and credit for technology demonstrations
(e.g., treadle pumps)
ƒ marketing assistance for business-oriented enterprises
ƒ communication and coordination of outgrower schemes
Although already very diverse, this list does not include some of the new services
recently taken on or proposed by the project, for example livestock health
services, forestry management, or conservation farming.
It seems unlikely that there is any organization in the project area, government or
private, that could fill the shoes of LFSP and provide such a diverse array of
services. On the other hand, some of these services may very well become
redundant or obsolete. For example, the identification of improved seed varieties
LFSP Mid-Term Evaluation ARD
103
for the three agroecological zones has already been done, and many AMCs have
already achieved a level of self-sufficiency in management of the seed
multiplication scheme, including procuring new batches of seed. Similarly,
training on CBO structures and certain IGA skills also does not need to be
constantly repeated, except when expanding to new areas. Other services currently
provided by the project may be replaced by the CBOs themselves, provided they
can empowered to draw-down required services from public or private sector
institutions. For example sourcing spare parts for a borehole or treadle pump is
something CBOs could learn to manage.
However some of the services provided by LFSP will continue to be in demand,
particularly for newer groups but also for ongoing groups. For example there will
mostly be a continued need for coordination with agriprocessors for outgrower
schemes. There will also most likely be a strong demand for the foreseeable future
for training in improved farming practices, as most of these technologies are still
in the demonstration phase. Most AMCs will also need financial and technical
assistance to reap the benefits of improved water harvesting structures.
Role of MAFF in Post-LFSP
At first glance MAFF would appear to be the institution best positioned to fill
many of the roles currently undertaken by the project. Although MAFF doesn't
have a mandate or resources for supporting the entire portfolio of services
provided by LFSP, for example income generation activities, proponents for a
hand-off to MAFF note that MAFF is the only institution with a large field staff in
the LFSP area and other rural areas. They also note that MAFF has a strong
agricultural focus, and provides solid technical training to its extension officers.
MAFF also possesses longevity and a legal mandate to serve the needs of rural
farmers.
On the other hand, others have questioned the capacity of MAFF to serve the
needs of rural communities at a level comparable to LFSP. This school of thought
notes that the traditional training and institutional culture of MAFF at the district
level does not encourage working through CBOs, despite the conducive national
and policy environment. They also note that even if training could adequately
reorient MAFF officers to embrace participatory approaches, constraints of
transport, salary incentives, and extension resources will always hamper the
effectiveness of MAFF's field operations. These constraints are deeply embedded
in the institution and are beyond the scope of donor funded projects to address.
This school of thought concludes by noting that LFSP does not have a particularly
good track record working with MAFF to date (see section 6.1.3 on page 33), and
that the history of other projects in Zambia and similar countries handing over to
government extension programs does not inspire confidence.
Private Sector
Like MAFF, private sector institutions may be able to fill some of the roles of
LSFP, but certainly not all. To begin, it is widely felt that it is in the economic
interests of agriprocessors in Zambia to develop stronger relationships with small
scale farmers, because Zambia does not have the volume of commercial farms to
LFSP Mid-Term Evaluation ARD
104
meet demand. So there is an incentive for private sector agriprocessors to support
mutually beneficial relationships with LFSP farmers. Larger agriprocessors have
field staff which can provide extension services to smallholder farmers. However
the cost of such services is eventually passed to the farmer, so it isn't cost effective
until a certain economy of scale is reached. Although private sectors institutions
benefit from the lower transaction costs associated with dealing with CBOs, the
private sector is not in a strong position to form new CBOs, or provide required
foundation training in areas such as CBO leadership roles, intervention
prioritizing, and record keeping.
Other Government Departments
Other government departments which provide services that impact food security
include the Veterinary Department (under MAFF), Forestry, and Social
Services/Community Development. These departments have similar constraints as
MAFF (e.g., transport resources, staff incentives), but they also don't have the
field presence that the MAFF extension program has. They do have however
specific technical skills and resources, and are potentially valuable support
partners for CBOs in the long term.
Other NGOs
Although LFSP is probably the largest NGO operating in Kazungula and Kalomo
Districts, there are other NGOs working in these districts. Two of the main ones
include World Vision and UNICEF, which focus on water, sanitation, and health.
In addition to the NGOs, there are some active charity organizations based at
churches. These groups for the most part don't use the broad-based livelihoods
approach that LFSP promotes, however they have expertise and resources in
specific activity areas.
Institutional Gaps
If LFSP were to pull out tomorrow, one would expect to see some of the
communities demonstrating self-sufficiency in some of the project activities, such
as the seed scheme. Other project services, such as facilitating planning for
watershed management, might not be within the grasp of current CBOs, but could
attained by those CBOs who have been empowered to draw down resources from
appropriate institutions.
However there are some programmatic areas where LFSP is really the only
adequate service provider in the area. Perhaps most important of these is CBO
capacity building. Building the capacity of CBOs sets the context for all other
activities and interventions. The CBOs are the focus of extension services,
marketing, and credit programs. They are central to LFSP strategy of improving
food security, and have their own spin-off benefits. Although most of the current
CBOs seem well on their way to becoming a least partially self-sustaining, it is
difficult to see how new CBOs could be formed or supported without LFSP under
the current institutional capacities. Marketing, which is becoming increasingly
important as farmers learn to increase production above subsistence levels, is
another area where there is no strong capacity outside of LFSP. Even within
LFSP, marketing currently seems to be on a piecemeal case-by-case basis. This
LFSP Mid-Term Evaluation ARD
105
type of marketing may be fine for small enterprises, such as dry season vegetable
gardens, however larger business initiatives like an outgrower scheme require a
stronger institutional marketing capacity.
Post LFSP-II Scenarios
The following scenarios try to illustrate some of the options available for
continuing and expanding the impact of LFSP past the project completion date.
These scenarios are highly simplified and represent different ends of the spectrum.
They are not mutually exclusive and are presented primarily to stimulate thought
and dialog between stakeholders.
Scenario 1: LFSP pulls out completely, with no concerted effort to replace or
sustain achievements.
Description: In this scenario, LFSP basically picks up and leaves the project area
at the end of the current funding period, without any kind of effort to hand
over activities to other institutions. If this were to happen, the long-term
impact of the project's achievements would largely depend on the
sustainability of the existing structures. The seed multiplication scheme and
use of low-rainfall crop varieties would probably continue, particularly in the
areas where it is already established, however there may be long-term
concerns of seed quality and supply. The impact of the water projects would
hopefully continue, assuming the beneficiaries could continue maintenance.
However it would be difficult for communities to construct or rehabilitate new
structures without any kind of external assistance. Small-scale income
generation activities would probably continue, for example linkages between
specific vegetable production interest groups and urban traders. However the
larger-scale IGAs which require more coordination, such as the Omnia
outgrower scheme, may be at risk. Activities which are strongly project driven
or supported, such as the monitoring program, would probably be gradually
neglected and eventually cease.
Strengths: Puts the sustainability of CBO approach to the test.
Weaknesses: Continuation of some LFSP accomplishments would be jeopardized.
Replication and expansion of project achievements would be slow and
uncoordinated at best.
Scenario 2: LFSP pulls out completely, handing over each of the four action
lines to an appropriate government department.
Description: In this scenario at the opposite end of the spectrum, LFSP still
completely pulls out all its resources, but passes responsibility for the different
activities to appropriate government departments. This might take place after a
certain transition phase (for example one year after the current funding period
ends) during which capacity building activities such as training take place, and
perhaps project equipment such a vehicles are transferred.
This is the "classic" post-donor scenario, where government is passed the
torch with the mandate to sustain and expand project achievements. MAFF
would probably take the lead role in the provision of services related to
farming systems. MAFF would probably also play a large role in CBO
LFSP Mid-Term Evaluation ARD
106
capacity building, as they are the only government department with a large
extension presence in the field. The Water and Sanitation, Forestry, and the
Veterinary departments would step into their respective roles supporting water
projects, NRM, and livestock health. Support for small economic activities
would probably be divided among multiple departments depending on the
focus of the IGA.
The relationship between CBOs and government departments would hopefully
be more customer-driven than past extension programs. Many of the existing
CBO-managed activities (such as the seed scheme) and linkages between
CBOs and the private sector would hopefully continue, but not be supported
except through the government departments.
Strengths: Government departments can be attractive handing-over partners as
they have a long term presence in the district and mandates to achieve
comparable goals. This approach can also have multiplier effects as
government departments offer mechanisms to expand the use of lessons
learned to other areas of Zambia.
Weaknesses: This history of the traditional handing-over approach to government
departments is not encouraging. The challenges are many. Government
departments have a tendency to not work well together, due to a variety of
bureaucratic, political, budgetary, and personal factors. The result can be an
uncoordinated and inadequate approach to service delivery, which loses the
synergistic benefits of a coordinated approach including higher cost￾effectiveness of implementation. Government departments also have some
significant operational constraints, including the civil service pay scale which
is not highly motivating, and limited resources for transport and
implementation of project activities. In some cases, government departments
can be highly politicized, making service provision inequitable and
inconsistent. So even in the best-case scenario, where the various department
heads completely understand and embrace the lessons learned from LFSP,
these constraints could very well limit the effectiveness of government to
sustain and expand LFSP achievements in food security.
This is not to suggest that government has no role to play in sustaining the
achievements of LFSP. Indeed the government departments have many highly
valuable assets, including highly-trained technical staff, and a long-term legal
mandate to serve rural communities. However shifting the responsibility to
carry forward the achievements of LFSP onto government exclusively does
not seem terribly attractive.
Scenario 3: LFSP gets another five years of funding from USAID or another
donor, maintaining its current scope, autonomy, and overall strategy.
Description: This scenario essentially represents a continuation of the operational
status quo, with perhaps some minor reorientation of project activities to
address already identified issues of joint forestry management, livestock
health, and watershed management. Although partnership building might be a
stronger focus than it has been in LSFP-II, the project would basically retain is
present staffing, autonomy, and geographic focus.
LFSP Mid-Term Evaluation ARD
107
Strengths: Continuation and intensification of LFSP's impact would continue
under this scenario. It would also allow for more refinement and dissemination
of lessons learned in important technology areas. The additional time could
also be used to plan and pilot test a transition strategy.
Weaknesses: Preserving the existing status quo would postpone, but not eliminate,
the need to address the difficult issues of sustainability. Taking additional time
to plan and prepare for a transition stage would probably be helpful in
planning an exit strategy, but learning will be minimal until the transition
actually takes place. The project may also be inflicted by the 'business as
usual' syndrome, where innovation and creativity are stifled from a perception
that the course is already well mapped out. An extended period of funding
would also reduce project incentives to focus on the difficult work on forging
institutional partnerships with government and the private sector, and retain its
focus on communities which are in many ways easier to deal with and yield
more visible results.
Scenario 4: LFSP becomes a training institution
Description: Under this scenario, LFSP continues to receive some kind of donor
support but slims down and focuses on what it does best: training. The target
for this training could be AMCs, however a more strategic audience would be
implementation partners. Topics would focus on areas where LFSP has
developed some expertise, namely CBO formation and capacity building,
participatory extension methods, water projects, and NRM technologies. To be
cost-effective, the target audience would have to be drawn from a wider area,
such as all of Southern province or even all of Zambia.
Strengths: Focusing on training would capitalize on LFSP's strengths and
experiences. It would also provide a much needed service to Southern
Province and Zambia as a whole.
Weaknesses: Training is an important component of capacity building, but by
itself can not improve food security. To be effective, training requires follow￾up support, on-site supervision and monitoring, and in some cases equipment
like seed or construction materials. This principle applies to both communities
and institutional audiences. Training addresses some of the problems of
partner institutions, such as overall orientation and technical skills, however it
doesn't impact other limitations such as pay incentives or resources for
implementing activities.
It is also difficult to demonstrate impact from training, a constraint that might
make this option less attractive to today's investors. Training institutes tend to
have high overhead costs because of the expense to build/maintain facilities.
There is also the potential of some duplication of effort with other training
centers in Zambia.
Scenario 5: LFSP evolves into a semi-permanent institution, either as a separate
NGO or a government-donor-CBO managed trust.
Description: In this scenario, LFSP evolves into a more-or-less permanent
presence in Southern Province, by either becoming its own NGO or a trust
managed by multiple stakeholders. As a separate NGO, LFSP would maintain
LFSP Mid-Term Evaluation ARD
108
its current level of operational autonomy, but would need to attract its own
funding. It could continue to focus on a holistic approach to improving food
security, or may need to specialize in training or specific technologies.
As a trust, LFSP would be guided by a board of directors which are drawn
from its main stakeholders: CARE, MAFF, local government, the CBOs,
traditional authorities, etc. Its funding base could potentially be derived from a
combination of donor investments, government contributions, fees for specific
services such as training, levies from outgrower schemes, or membership fees
from CBOs. The size and direction of its activities would be determined by its
board of governors and resource base, but would probably focus on the
capacities developed by LFSP: CBO formation and capacity building,
improved farming systems, water harvesting, and income generation.
Strengths: A semi-permanent, semi-autonomous food security trust in Southern
Province would address many of the issues concerning sustainability of project
impact. Participation from a wider range of stakeholders could potentially
improve dissemination of lessons learned and strengthen institutional
commitment to the participatory approaches which are at the core of LFSP.
Coordinating multiple services through a single institution would maintain the
holistic livelihood approach to food security that LFSP has promoted, improve
cost-effectiveness of service delivery, and maximize synergistic effects. This
type of structure would also have greater longevity than a purely donor-funded
intervention. There is also some indication that government has recognized the
value of a trust. MAFF is currently a partner in at least two trusts, the Cotton
Development Trust and the Golden Valley Agricultural Research Trust (see
box below). The Government's Livestock Development Plan also floats the
idea that livestock services can be delivered through semi-autonomous trusts.
Weaknesses: Although there is some precedent of trusts in the agriculture sector in
Zambia, experiences with this type of partnership for servicing smallholder
farmers are minimal. Numerous challenges line the road ahead. First and
foremost is cost. LSFP has been able to achieve results by maintaining a staff
of approximately 50 in two districts, with an annual operating budget of
approximately $800,000. Achieving even half this scale in a trust would be
challenging without donor subsidization. A trust may also be challenged to
stay focused. Food security is a broad area and a large array of interventions
can be implemented in the name of food security. A diverse Board of
Directors representing multiple institutions is definitely an asset in many
ways, but could also be a liability when it comes to defining program
direction.
An Multi-Stakeholder Ag-Sector Trust Model To Look At
The Golden Valley Agricultural Research Trust (GART)
GART was established in 1994 and became operational in
September 1995. It is directed by a Board of Trustees representing
the Zambia National Farmers Union, MAFF, University of Zambia,
ICRISAT, and a financial institution. The Trustees provide policy
direction while a Management Board carries out the annual work
program. The Management Board is backed up by four technical
and advisory committees drawn from a broad range of stakeholders.
LFSP Mid-Term Evaluation ARD
109
The main activities of GART are contract research, service
research, and commercial cropping and ranching. It is funded
through its contract research and commercial cropping operations,
with additional support from the member institutions and World
Bank (through MAFF). They publish an annual report called the
Year Book which summarizes the activities of the Trust and
presents research results.
GART has proven effective in partnering private and public sector
institutions, using agricultural research as a common focus. The
representation of ZNFU on the board of directors and the financial
dependency on contract research ensures that GART's services
remain demand-driven. Although the focus on research for
commercial agriculture and broad financial base are a far cry from
a trust that would be centered around smallholder farmers, GART
presents an interesting partnership model that combines the
strategic interests and needs of multiple stakeholders to provide a
much needed service in the agriculture sector.
Scenario 6: CARE opens a new food security program in a neighboring
district4
.
Description: In this scenario, LFSP applies the same basic approach it has used in
Livingstone/Kazungula/Kalomo districts to another area.
Strengths: This option would allow CARE to apply the lessons learned from LFSP
in a new context. This type of experimentation is an extremely valuable
exercise, because it allows projects to differentiate which approaches are
context-specific and which have broader application. Expansion to new areas
would also provide an important service to the people of Zambia, as the
approaches developed by LFSP are designed to work in marginal areas where
food security is a major concern. There are a number of such marginal areas in
Southern Province as well as other parts of Zambia including Western and
Central Province.
Weaknesses: The sustainability issues and need for an exit strategy for the LFSP
areas would still need to be dealt with, and CARE would have to examine its
capacity to implement a second food security project in Zambia. The project
has made a strong investment in staff development, and would need to find a
way to reallocate these resources to support a new area. Donors may also be
hesitant to fund a new food security project without first seeing the long-term
sustainability of the parent project.
Conclusion
As LFSP enters its final year of USAID financing, it needs to begin serious
dialogue internally and externally to plan its next move. The discussion and
 
4
 This option doesn't really offer much to the dialog on the post-LFSP transition, because it doesn't
present an option for an exit strategy in the existing project area. However it is presented here because
it has been raised by a number of project and CARE staff.
LFSP Mid-Term Evaluation ARD
110
scenarios presented here only scratch the surface of the thought process that needs
to go into the preparation of new proposal. Each scenario presented here could
easily be flushed out into 30 page proposal or more, outlining a rationale and
operational framework. Some of the scenarios above are clearly not very attractive
and only presented for discussion purposes. Perhaps the best approach will
represent a combination of several of the above scenarios or other creative
solutions.
The evaluation team does not feel it is in a position to recommend one of the
above strategies above the others, however it does recommend a general approach
which is summarized below.
Evaluation Team's Recommendations:
Five-Step Process for Preparing for Life After LFSP-II
1. CARE/Zambia should reexamine and rearticulate its strategic
and manageable interests in food security.
2. LFSP should articulate the comparative advantages of its
livelihood approach, both as a whole and for each of the four
action lines.
3. CARE/LFSP should articulate their goals for post LFSP-II.
4. CARE/LFSP should initiate private discussions with potential
institutional partners on post-LFSP, to be followed by a
broader workshop once a tentative framework is developed.†
5. CARE/LFSP should conduct or contract the necessary
feasibility studies, particularly if there is a proposal to expand
to new areas or evolve into a semi-autonomous NGO or trust.
†
this could be incorporated into the agenda of the ongoing series of
partnership workshops
LFSP Mid-Term Evaluation ARD
111
11.0 LESSONS LEARNED
The following lessons have been learnt in the process of implementing LFSP. These
lessons could be used for improving project planning and effectiveness in future.
11.1 LFSP Lessons Learned
CBO Structures and Capacity Building
1. CBOs can learn how to use PRA methods in their own work within their
communities. Some AMCs and VMCs have for instance done PRA work with
neighboring villages to help them launch seed programs.
2. The peer pressure of the cell group is valuable in enhancing the performance
of members and contributes to achieving targets set by the communities
themselves. This has been demonstrated in loan schemes and construction
work.
3. Although cell groups and the whole CBC structure appear to be sustainable
mechanisms that have not faded away as the novelty of the LFSP wore off,
group durability is not necessarily the best criterion for measuring the success
of the LFSP extension approach. Groups should only last for as long as people
find them useful. Evolution of groups is inevitable and usually healthy.
4. PRA processes may create expectations among rural communities that may be
outside the purview of the project. Care must therefore be taken to be realistic
about how soon action will be taken as it leads to resentments when nothing
happens.
5. Once people are familiar and comfortable with the experience of working in a
CBO, they can organize themselves to work together in various development
initiatives.
6. LFSP experience shows that flexibility about local institutional structure is
empowering for local people and promotes their active collaboration with the
project.
7. It is important to include traditional authorities (Chiefs and Headmen) in early
contacts and briefings in order to avoid clashes between them and the CBO
structure developed under the project in order to facilitate collaboration.
8. Mediation and conflict resolution mechanisms are required in CBO structures.
Under LFSP there are cases in which VMCs and AMCs have used headmen
and chiefs for this purpose.
9. Some CBO leaders in the LFSP areas have successfully ran for local political
offices (District council) because of their high profile in the community.
LFSP Mid-Term Evaluation ARD
112
10. LFSP’s experience shows that poor rural people will respond positively and
competently to a clearly formulated development opportunity that meets a
plain and widely shared development need.
Farming Systems
11. A community based seed multiplication scheme of early maturing and
drought-resistant crops is an effective way to alleviate food insecurity in
drought prone areas.
12. Local seed banks help ensure that seed is available when needed, and that it
won't be eaten before planting in times of hunger.
13. Successful interventions in increased production must be supplemented with
training on post-harvest technologies for food storage and processing.
Increased production also results in a demand for improved crop marketing.
14. It is possible to increase land productivity considerably using soil
improvement crops such as velvet beans and sunhemp.
15. Factors that could limit adoption of green manuring techniques include lack of
enough land to allow for fallow, labor constraints, lack of equipment to
incorporate the green manure into the soil and the fact that farmers find it hard
to grow a non-food crop.
Gender
16. LFSP has learnt that reducing gender inequities in rural Zambian society is a
delicate, long-term challenge. Achieving more equitable numerical gender
balance in areas such as CBO leadership positions does not necessarily lead to
any immediate fundamental shift in gender relations, although it may be the
first step in meeting the longer-term challenge.
17. One method of empowering women is to raise their income by promoting
crops traditionally grown by women.
Sustainability & Expansion
18. Although CBO dependency on the project is still a concern, there is evidence
that AMCs can operate autonomously, taking their own initiatives and linking
themselves to government and other NGO agencies. There are, however, still
many issues on which CBOs still expect help from LFSP.
19. Program expansion into new areas should be based on interest expressed by
household and the communities, and not on rigid pre-determined processes.
20. Increasing food production is a major step which bridges relief and
development, both within the community, as well as for the NGO involved.
Once basic food needs are met, a wide array of other social needs will be
voiced which may over-stretch the capabilities of the project.
LFSP Mid-Term Evaluation ARD
113
Monitoring and Evaluation
21. CSM ledgers are an effective mechanism to collect and manage household
level data on demography, household assets, and production. However
teaching people how to use information for their own planning purposes has
proven to be a challenge.
22. CSM works best as tool for collecting factual information for purposes such as
identifying households in greatest need of food relief, or checking which crops
are most popular or successful. Broader identification of trends and issues has
mainly emerged from less structured discussions within CBOs and between
them and the project.
Extension
23. Start-up costs for CBO extension are comparatively high, however running
costs of the extension approach are comparatively low. Overall extension
through CBOs is more cost-effective than T&V.
24. By providing extension services through CBOs and taking advantage of
community facilitators, one extension officer is able to work with
approximately 1,000 farmers.
25. Extension services can be provided by NGOs equally as well as by
government extension programs.
26. It is possible for an NGO extension program to help influence the approach
used by a government extension program, through documentation, training
and examples.
27. Experiential learning and farmer-to-farmer extension approaches are effective
in spreading conservation farming ideas.
28. Local facilitators have shown themselves to be quite capable of providing
extension services in their communities. Use of "local" technologies, such as
live fencing using a local tree species, is much more intuitive for facilitators to
share with other farmers. Introduction of technologies which are not well
known, such as green manure crops, requires more training before facilitators
can effectively disseminate the technology.
Partnerships
29. Forming a durable partnership requires making an investment in working with
the other institution from the very beginning.
30. When there is no durable, structured agreement between two institutions such
as LFSP and MAFF, then the relationship is going to be inconsistent,
opportunistic, and largely defined for better or worse by personalities
involved.
Private Sector Linkages
LFSP Mid-Term Evaluation ARD
114
31. Marketing strategies enable rural households to get better prices when they
sell by increasing volume, reducing uncertainty, and decreasing transaction
costs. By developing proper linkages with established traders, community
members get a better bargain for the produce.
32. When demonstrating new technologies that require private sector services
(e.g., parts, training) to establish/maintain, it is better to get the private sector
partners involved from the very start, (e.g., let them do the training and
installation) so that working relationships are built with the CBOs.
Water Harvesting
33. Water for household use has been achieved for some areas but distances are
still large for some households.
34. There seems to be low utilization for some of the water resources, for example
in fish farming and gardening. Because of uncertainty about rainfall,
communities tend to limit the utilization of water basically for household
consumption and livestock.
NRM
35. Initial establishment of NRM activities is time consuming. However, if
benefits can be made tangible, appropriate technologies stand a better chance
of being rapidly disseminated. For this reason it helps to target areas where
benefits will be seen quickly such as infertile fields and silted dams.
36. Participatory monitoring programs help demonstrate the long-term benefits of
NRM.
Income Generation
37. The chronic risk of drought in Southern Province highlights the need for
strengthening the asset base of households as a coping mechanism against
food shortages. Strengthening the asset base requires increased revenue.
38. Savings, credit and marketing, among rural communities are very difficult
interventions in which to make progress. They require intensive and extended
professional support in order for commercially viable results to be achieved.
Savings programs are not an effective strategy in an economic environment
where bank interest is lower than inflation.
11.2 Expanding Lessons Learned
Although there is no magic recipe for expanding or replicating the lessons learned
from a project, there is a general process that replication follows (see Figure 67).
It begins by documenting the lessons learned and disseminating the results. This
can be achieved by several strategies including traditional reports, presentations at
conferences and meetings, web sites and email lists, newsletters, the electronic
and print mass media, outreach, professional associations, etc.
LFSP Mid-Term Evaluation ARD
115
Once lessons learned are disseminated, the next step is for the leadership of other
institutions to take notice and incorporate the lessons into their own institutional
memory. This step is sometimes forgotten by projects, who sometimes believe
that traditional reporting formats will be so compelling that other institutions will
start knocking down doors to find out more about the effective strategies. Projects
that have a genuine commitment to maximizing the yield of their investments in
testing new approaches will explore strategies such as a user-friendly newsletter to
make sure that their lessons are presented compellingly and repeatedly to key
audiences.
After another institution becomes interested in the lesson, they will review it in
greater depth, test, and adapt the approach or technology to the their own goals
and context. In the case of projects the new tool/method can be incorporated into
project activities. Government departments can integrate lessons learned into their
operational frameworks and policy, while donors can integrate lessons into future
cooperative agreements and the design of portfolios.
LFSP Mid-Term Evaluation ARD
116
Lessons Learned
Documented
Experience
Monitoring Research
Information
Shared
Project A
Takes Notice
Replication
Government
Takes Notice
Review
Testing
Adaptation
Integrated into
Operational
Framework
Policy
Change
Instutional
Memory
Instutional
Memory
Instutional
Memory
Exchange
Donor
Takes Notice
Review
Cooperative
Agreements
Portfolio
Design
Testing
Adaptation
Project A
meets Project B
*
*
* link where donors can provide assistance
Figure 67. A framework for expanding lessons learned.
A donor or project can only do so much to encourage the replication of their
lessons learned, but they are in a position to impact certain stages of the process,
primarily in documentation and dissemination. To maximize the likelihood that
the appropriate people will take notice, lessons should be shared compellingly and
repeatedly using a variety of media.
LFSP Mid-Term Evaluation ARD
117
Figure 68. LFSP planning workshop
Inviting external stakeholders to planning workshops is one way to disseminate lessons learned
which can lead to replication
LFSP Mid-Term Evaluation ARD
118
12.0 ADDITIONAL RESEARCH NEEDS
Like most research studies, this evaluation identified additional research topics that
would help to elucidate the impact that LFSP has had on food security and rural
income. These topics were either not explicit in the scope of work or were beyond the
time available during the evaluation period. They are presented here to help guide
future research planning and to give some ideas which may help the project prepare
for its final evaluation.
12.1 Return on the donor investment
From the donor perspective, a more interesting question than how much USAID
financing was spent, is how much return was delivered for each dollar invested
into LFSP. Such an exercise for LFSP would be complex to say the least, as the
project delivers a multiple of services, many of which can not be easily translated
into monetary terms. CBO capacity, improved health, business skills, etc. have
economic value in the sense that they contribute to increased agricultural
production, however they also have intrinsic value which is not easily quantified.
But at least on a section by section basis, the project may be able to develop a
methodology and sampling scheme to estimate the return on the investment dollar.
12.2 Project impact on household asset base
Although this evaluation made some progress in measuring the impact of LFSP on
the asset base of participating households by using information from the CSM,
there wasn't enough data examined to make a definitive conclusion. The project
has also conducted some studies on household assets, such as the increase in
livestock, but there is need for a more systematically conducted analysis. Another
unknown question is the differential impact of the project's interventions of the
different wealth categories. In other words, do the poor make more or less
progress than the rich? This line of inquiry would be very pertinent for the final
evaluation.
12.3 Institutional capacity of partner institutions
Although the scope of work for this evaluation called for an assessment of the
capacity of partner institutions, time did not allow for an adequate analysis to be
made. This question will be highly relevant for planning LFSP Phase III, whatever
shape it may take (see 10.5 on page 101). Questions such as staffing levels and
training, transportation and communication resources, planning capabilities,
strategic orientation, etc. will help shape the dialogue leading to LFSP's next step.
12.4 Accuracy of Community Self-Monitoring Data
Although the agricultural technologies get most of the attention in LFSP, the use
of the Community Self-Monitoring system is one of the most innovative
implementation tools developed by the project. This type of method is also highly
transferable to other development arenas. However the validity and accuracy of
the data in the CSM has not been examined in great detail. Both LFSP and the
LFSP Mid-Term Evaluation ARD
119
field of development in general would greatly benefit from a study on the
accuracy of this approach. Questions include:
ƒ How complete is the data in the CSMs?
ƒ How accurate is the data in the CSMs? What are the biases?
ƒ When is the CSM data collected and recorded? How long does it take?
ƒ How much of the CSM data is based on farmer recall, and for how far
back?
ƒ How much production data is missed due to crops being eaten before
harvest?
ƒ How accurate are the measures of area and mass in the CSM?
ƒ How consistent are the measures of area and mass in the CSM?
ƒ What are the incentives for maintaining the CSM?
ƒ What is the minimum amount of external support and supervision required
to keep the CSM going?
With more detailed data on the CSM such as the questions above, the project
would be able to make a stronger assessment on the validity of the data. They
could start to put confidence intervals around averages (e.g., "an average of 0.2
±0.04 ha of MMV400 was planted in year X"), which would greatly strengthen
the interpretation of results, and provide a very useful tool for the final project
evaluation.
12.5 Impact of LFSP on Food Relief
The underlying attraction for LFSP as an investment option is the argument that it
is cheaper to address the underlying causes of food security than it is to provide
food relief. Although Southern Province has not experienced the severe droughts
that it did during the first half of the 1990s, there has been some food relief in the
province since the project started. If the project is achieving its expected goals, we
would expect to see less food relief needed in the project area than in non￾participating areas, and also less food relief needed than in comparable droughts
before the project began. There is also preliminary evidence that when food relief
is required in the project area, it is distributed much more efficiently and
effectively when CBO structures are in place. Such a study would not be terribly
difficult to conduct, and may present very compelling evidence for other donors
involved in food relief to adopt approaches exemplified by LFSP.
12.6 Impact of Technologies on Production
Although LFSP recently completed a production trends study, there are still some
questions on the impact of technologies that need to be addressed in more detail.
The question of changes in seed quality and the impact of seed quality on
production has been examined on a small scale, however a robust analysis which
includes other factors such as rainfall and farming practice is still needed.
Likewise the impact of soil conservation and soil fertility technologies on
production has been studied on a small scale, but more examples would be
LFSP Mid-Term Evaluation ARD
120
helpful. Finally the adoption rate of the improved farming practices, both for
participating and non-participating factors, is not well documented.
The project has begun to take steps to collect more data on production, including
crop-cuts planned for 2000, which will provide an excellent data source for a more
robust analysis of production. The Food Production Trends Survey can also
provide relevant data to show changes in production over a four year period. This
type of study would also help the project quantify its overall impact on food
security in Southern Province, something it would be challenged to produce today.
LFSP Mid-Term Evaluation ARD
121
13.0 RECOMMENDATIONS
Overall the evaluation team was impressed with the performance results of LFSP. It is
a well-conceived and managed project, and has an innovative experimental focus not
found in many development projects. It may still be a bit too 'top-down' in its
approach to providing services to communities, and faces serious challenges in
developing mechanisms to achieve long term sustainability, but overall the evaluation
team feels this project has been a good investment for USAID/Zambia and is worthy
of continued support.
13.1 Recommendations USAID's Future Investment Options
The matrix on the following page summarizes the advantages and disadvantages
of USAID's choices for future investment in LFSP. The evaluation team
recommends the following elements of the matrix:
1. LFSP should not expand its geographic scope into neighboring districts at
this time because it does not have the required administrative or technical
resources
2. LFSP should expand its target population in the existing areas by targeting
new AMCs, in order to capitalize on the benefits of scale without
significant additional overhead on project resources
3. LFSP should be encouraged to expand its scope of technologies in the
areas already identified by the project, namely livestock health, and NRM.
4. USAID should continue to support LFSP and encourage the project's
efforts to develop an exit strategy which maximizes the likelihood that the
project's achievements will be sustained
122
Table 14. Advantages and Disadvantages of USAID's Investment Options in LFSP
Investment Option Advantages Disadvantages
Extension • continue making progress towards SO1
• provide sufficient time to fully develop all components of the
LFSP strategy and determine if the model works
• provide additional time for new technologies and intervention
strategies to be developed and be studied
• see the benefits of long-term interventions such as NRM
technologies
• better opportunity for replicating best parts of the model
• phase out can be gradual and deliberate instead of rushed and
improvised
• opportunity for maximizing contributing support from other
NGOs and donors
• USAID can build upon past investments rather than invest in
developing a new project
• additional funding needed
• reduced incentive for project to begin
phasing out process
Expansion:
Expansion into new
districts
• increase progress towards SO1
• opportunity to capitalize on existing knowledge gained by the
project
• opportunity to further test the social and ecological
conditions/ assumptions upon which the LFSP model is based
• multiplier effects from new experiences/ exposure of model
• opportunity to test new strategies of establishing CBOs &
project/community relationships
• parts of model not yet fine tuned and not yet
ready for replication
• project doesn’t currently have management
and administrative capacity to serve new
areas
• expansion may detract resources from
services in existing areas
• additional time and money required
Introducing program
into new communities in
existing districts
• basic project infrastructure is already in place - increase
impact with minimal additional resources
• benefits of scale:
• additional funding needed
• project staff may become overtaxed
• components of program that don’t work
123
Investment Option Advantages Disadvantages
• a larger critical mass might accelerate adoption of new
technologies
• marketing advantages
• inter-community synergism
• new possibilities for institutional partnership
• reduces unit cost for services delivery
• opportunity to test new methods of establishing CBOs and
defining a CBO-project relationship based on lessons learned
• improve achievement of project goals - greater impact per
capita in the three districts
well might not be ready to implement on a
larger scale
Expand scope of
technologies and
strategies (e.g., livestock
health, NRM, CF,
donkeys, information
dissemination)
• preliminary evidence suggests good potential for new
strategies
• able to study synergistic effects of multiple interventions
• more experimentation will result in a stronger intervention
model
• increased opportunity for developing partnerships - multiplier
effect
• requires additional time and funding
• new strategies may not work
• new strategies may not contribute much to
project goals or SO1
• if new intervention strategies are not
demand driven then adoption may be low
• more direct services may increase donor
dependency syndrome
• possible loss of focus on core project
activities
Cutting short • SO1 funding can be made available for other programs
• opportunity to study the consequences of donor pull out
• communities not ready to function on their
own
• lose institutional structure and human
knowledge to plan and implement future
food security interventions
• if project activities fail to be sustainable,
there is a likelihood that increased food
relief will be needed
• loss of contribution towards SO1 goals
124
Investment Option Advantages Disadvantages
• monitoring likely to cease, lost learning
opportunity to see if model works
• lost opportunity to finish developing and
testing the LFSP intervention strategies
• loss of donor credibility by the beneficiaries
and institutional partners, reinforcing the
perception donors have limited short-term
interests
LFSP Mid-Term Evaluation ARD
125
13.2 Programmatic Recommendations
In addition the general recommendations listed above, the evaluation team makes
the following programmatic suggestions:
1. Develop activity-oriented MOUs with local government institutions and
NGOs.
2. Develop an institutional strategy for expanding the delivery of demonstrated
technologies through linkages with private sector and government institutions.
3. Revisit the performance indicators, particularly those of the SEAD section, in
light of the recent reorientation of the operational framework (e.g., stronger
emphasis on marketing, livestock health, partnering, AMC graduation, etc.)
4. Strengthen monitoring of SEAD activities. Data is there but is not organized.
This should be a focus of the new SEAD coordinator when the position is
filled. A basic set of variables (e.g., number of beneficiaries, gross and net
income) should be standardized across all interest groups so generating impact
data will not be difficult. Consider the CLUSA RGBP Profit & Loss
statements as a possible model.
5. Begin compiling electronic copies of all documents for eventual distribution
on CD ROM or the web, as a cost-effective dissemination mechanism which
can supplement the distribution of hard copies.
6. Strengthen the methodology section of project reporting, including a
description of the sampling methods and possible biases.
7. Strengthen monitoring of technology trials, including a greater number of
demonstrations and generation of real production data for improved farming
systems and NRM technologies.
8. Make better use of CSM data for impact monitoring at the project level.
Strengthen training of CSM introducing more accurate and standardized
measurements for area planted, types of assets recorded, and production units.
Add variables on the trade or sale of crops.
9. Strengthen the collaboration with other technology focused agriculture
research programs, including CFU and ICRAF.
10. Strengthen the information technology resources of the project, in particular
the Monitoring and Evaluation section, either by building internal capacity or
by establishing a long-term tech support contract with an appropriate firm.
11. Increase the amount of collaboration between MAFF and LFSP extension
officers before handing off graduated CBOs to MAFF. Allow sufficient time
LFSP Mid-Term Evaluation ARD
126
for extension officers to work side by side to become oriented and comfortable
with participatory methods of extension and facilitation.
12. Work closer with local traditional authorities, especially in the areas of NRM
and joint forestry management. Be mindful of the experiences of other projects
such as ADMADE in regards to strategies for working with traditional rulers.
LFSP Mid-Term Evaluation ARD
127
ANNEX I - PEOPLE CONTACTED
USAID/Zambia
P.O. Box 32481, Lusaka
Tel: 260-1-254303
Fax: 260-1-254532
David Soroko Agricultural Development Officer
Morse Nanchengwa Agriculture and Natural Resources Program Officer
Chris M. Muyunda Small Enterprise Development and Agri-Business
Program Officer
Helen K. Gunther Agriculture and Natural Resources Development
Officer, USAID/Madagascar
CARE International in Zambia
P.O. Box 36238, Lusaka
Tel: 260-01-265901
Email: care-zambia@zm.care.org
Mark VanderVort Assistant Country Director
Godfrey A. Mitti Head-Agriculture and Natural Resources
Colin Guanu Director of Finance
Regis Gwaba Monitoring and Evaluation Unit
Golden Valley Agricultural Research Trust
P.O. Box 50834, Lusaka
Tel: 260-1-611150, 611018, 611163
Fax: 260-1-611232
Stephen Muliokela Director
Douglas Moono Senior Agronomist
Sosten Lungu Assistant Agronomist
Others - Lusaka
A. K. Banda Acting Permanent Secretary, MAFF
Robert Chimambo Ministry of Water Affairs
Chansa Mushinge FEWS Representative in Zambia
Paul Muzunda AFRICARE/Zambia
Dr. Lungu UNZA, Soil Science Department
Alex Mwanakasale Agricultural programs, World Bank
Charles Chileya National Programmes Coordinator, FAO
Livingstone Food Security Project
P.O. Box 60256; Livingstone
Tel: 260-03-323244, 323730, 324259
Fax: 260-03-320687
Email: clivings@zamnet.zm
LFSP Mid-Term Evaluation ARD
128
Sylvester Kalonge Project Manager
Robby Mwiinga Assistant Project Manager
Samson Mbale Technical Coordinator
Jasper Hatwiinda Monitoring and Evaluation Coordinator
Catherine Pongolani Farming Systems Coordinator
Florence Chawelwa Administrative Coordinator
Helen Khunga Extension Officer
Sara Mwale Extension Officer
Musonda Kapena Extension Officer
Highland Hamududu Training Officer
Alice Whitehead PAEC Assistant
Fungaloko Patson SEAD Officer
Felix Lombe Technical Coordinator
Nathaniel Mkuzo Water Technician
Fred Sichone Senior Accountant
Christopher Kakunta Documentalist
Evans Mwengwe Livestock Officer
Agnes Bwalya Secretary
Beneficiaries - Livingstone
Patson Sibibi Songwe AMC facilitator, Mupotola Village
Daniel Akoyaka Member, Songwe AMC, Mupotola Village
Dragon Zondani Member, Songwe AMC, Mupotola Village
Dean Mullwena Secretary, Songwe AMC, Mupotola Village
Ben Sialutaba Chairman, Songwe AMC, Mupotola Village
Jack Mwamba Jack Mwanampapa AMC, facilitator
Mrs. Jack Mwamba Member, VMC
Bobby Mukuni Member, VMC
John Mwamba Member, Jack Mwanampapa AMC
Mrs. John Mwamba Member, VMC
Forestry Staff LFSP
Mr. Ngenda Delevu AMC Chairman, Advisor to Chief Sekute,
Councilor
Sofia, Mrs. Ngenda Delevu
Ritious Mulena Makunka AMC facilitator, Simalolo Village
Mary Simunene Makunka AMC Vice-Seceatary, Simalolo Village
Caesar Sami Farmer & Sinde AMC Chairman, Mujala Village
Donald Hangombe Sinde AMC Secretary, Mujala Village
Donald Chiiko Member, Sinde AMC, Mujala Village
Audrey Simakuni Member, Mujala VMC, farmer, housewife
Mathews Syachitema Katapazi AMC facilitator, 5 villages, Katapazi Village
Oscar Musena Katapazi AMC Chairman, Katapazi Village
Kenan Simasiku Katapazi AMC facilitator, 4 villages, Katapazi Village
Ackim Syachitema Shopkeeper, Katapazi Village
Katapazi VMC Chairman, Katapazi Village
Micheal Siamujaye School teacher, Deputy Headmaster, Katapazi School
Vincent Ngumbili Shindu AMC Chairman, Shindu Village
Kenneth Likokoto Member, Shindu AMC, Shindu Vilalge
LFSP Mid-Term Evaluation ARD
129
Charles Rikokoto Member, Shindu 2 VMC
Samson Rikokoto Member, Shindu 2 VMC
Morgan Sikumba LFSP/Kalomo Office Helper
Linda Moonga LFSP/Kalomo Office Helper
Argent Hamutiki LFSP/Kalomo Driver
Mr. Sitali LFSP/Kalomo Tractor driver
Wilson Kafwanka LFSP/Kalomo Mechanic
Moses Wachila Mweemba AMC facilitator
Mr. Thompson
Wachila
Headman, Mweemba Village
Benny Simwambwa Kasukwe AMC Chairman
Renard Twaambo Member, Kasukwe AMC
Jacob Mudenda Kasukwe AMC
Akim Muzunga Member, Kasukwe AMC, farmer, Chikombi Village
Kerry Mungala Member, Chikombi VMC
Fanny Mooya Farmer, Chikombi Village
Clever Muleya Farmer, Chikombi Village
Miyoba Hanyuka Farmer, Chikombi Village
Josta Kaliilo Daughter to Akim Muzunga and Kerry Mungala
Mr. Monga Kasukwe Ward Councilor
Others - Southern Province
Mr. D.H. Hakayobe Permanent Secretary Southern Province
Mr. Banda Kazungula/Livingstone District Planning Officer
Christine Chisembele Kazungula District Veterinary Officer
Mwape Walumba Southern Province Social Welfare Department
Katupa Chongo Kazungula District Agricultural Coordinator
Joy Sinyangwe Crops Specialist, Kazungula District
Ms Malambo Kalomo District Agricultural Coordinator
M.P. Siyamasamu Social Welfare, Southern Province
K. Ngalande Kazungula District Community Development
Officer
Freddie Kamulosa Acting Forestry Officer, Kaloma District
Lumumba
Mwanangonze
MEDO/MAFF, Kalomo District
Misheck D. Zulu Roads Department, Kalomo District
L. Chileshe Senior Administration Officer, Kalomo District
Council
Mukombo Cheelo A/WDO, Kaloma District
Cruesha Muleya DCDO, Community Development, Kalomo
District
Mubita Musiwa Kalomo District
Martin Sampa Southern Province, Development Planning Officer
Willem Colenbrander Director, Busiku Consultants, Southern Province,
Livingstone
Silvasy Shibulo MAFF District Crop Husbandry Officer,
Livingstone
LFSP Mid-Term Evaluation ARD
130
ANNEX II - LITERATURE CONSULTED
CARE. (1996). Project Implementation Order/Technical Services: Livingstone Food
Security Project . Lusaka: USAID.
Drinkwater, M. (1997). Participation within participatory technology development
and dissemination: Zambian case study . Livingstone, Zambia: Livingstone Food
Security Project.
Hatwiinda, J., & Whitehead, A. (2000). Searching for sustainability: An impact
evaluation of the Livingstone Food Security Project training programme .
Livingstone, Zambia: Livingstone Food Security Project.
Kalonge, S., & Pongolani, C. (1997). Social and Economic Benefits of Community
Seed Bulking: A Case of the Livingstone Food Security Programme . Livingstone,
Zambia: Livingstone Food Security Project.
LFSP. (1998a). Annual report: July 1997 through June 1998 . Livingstone, Zambia:
CARE Zambia.
LFSP. (1998b). Quarterly report #09: October through December 1998 . Livingstone,
Zambia: Livingstone Food Security Project.
LFSP. (1998c). Report on Gender Stakeholders Workshop . Livingstone, Zambia:
LFSP.
LFSP. (1999a). Annual report: October 1998 through September 1999 . Livingstone,
Zambia: CARE Zambia.
LFSP. (1999b). Quarterly report #11: April 1999 through June 1999 . Livingstone,
Zambia: Livingstone Food Security Project.
LFSP. (1999c). Quarterly report #12: July through September 1999 . Livingstone,
Zambia: CARE Zambia.
LFSP. (1999d). Reorientation of a number of LFSP activities . Livingstone, Zambia:
Livingstone Food Security Project.
LFSP. (1999e). Seed Scheme Assessment Report . Livingstone, Zambia: Livingstone
Food Security Project.
LFSP. (2000). Quarterly report #14: January through March 2000 . Livingstone,
Zambia: CARE Zambia.
Lyons, A. (1998a). A Profile of the Communtity-Based Monitoring Systems of Three
Rural Development Projects in Zambia . Lusaka: USAID/Zambia.
Lyons, A. (1998b). Trip Report to the Livingstone Food Security Project . Lusaka:
USAID.
MAFF. (1999). Final Project Proposal of the GRZ/Sida Food Crop and Seed Research
Project . Lusaka, Zambia: Ministry of Agriculture Food and Fisheries.
MAFF. (2000). Pilot Participatory Extension Approach for Zambia . Lusaka, Zambia:
Department of Field Services, Ministry of Agriculture Food and Fisheries.
LFSP Mid-Term Evaluation ARD
131
MENR. (1999). National Biodiversity Strategy and Action Plan . Lusaka, Zambia:
Ministry of Environment and Natural Resources.
Milimo, J. T., Kasonde-Ng'andu, S., Pongolani, C., Zulu, A., & Mwanza, E. (1997).
Marketing Consultancy . Lusaka, Zambia: Livingstone Food Security Project.
Milimo, J. T., & Tripp, R. (1999). The diffusion of seed and information in the
Livingstone Food Security Project . Livingstone: International Crops Research
Institute for the Semi-Arid Tropics.
Mwiinga, R. (2000). Natural Resource Management Replication Strategy .
Livingstone, Zambia: Livingstone Food Security Project.
Ndiyoi, M., Bwembya, P., & Kasuta, E. (1999). Work Ends, Knowledge Endures:
Midterm Review of LFSP . Livingstone, Zambia: Livingstone Food Security
Project.
Turner, S. D. (2000). Extension that works: Lessons learned by the Livingstone Food
Security Project . Lusaka: CARE Zambia.
LFSP Mid-Term Evaluation ARD
132
ANNEX III - ENVISIONING THE FUTURE
The following fictional story tries to capture one vision of how LFSP AMCs may exist
ten years from now - after the project is gone. The use of hypothetical narratives such
as this one can be a useful tool to help crystallize and articulate various perceptions
of the ultimate goals of a project, a best-case scenario resulting from achievement of
all project goals. Once a destination is visualized, a path in the right direction can be
charted from the present location in the journey.
Livingstone, May 2010. James Miyanda feels a little nervous as he sits in the reception
area of the District Education Office, waiting to see the District Education Officer for
Kazangula. He is still relatively new as an AMC Chairman, having been elected just a
year ago. Before that he was Secretary of the AMC and Chairman of his VMC, where
he was responsible for record keeping for the community seed scheme. Although he
feels very comfortable working with the people in his community on activities such as
the seed bank, maintenance of the weir, and the new outgrower scheme, he has never
worked with anyone from the District Education office and only has a grade 12
education himself. He wonders if he'll be respected here but so far everyone has been
very helpful.
He has come to Livingstone, along with the facilitator for his AMC, to inquire about a
new service which he has heard about. They have heard that the education department
has a program whereby they can send an additional teacher to rural areas for girl
education. The issue of education for girls has been coming up in meetings at the
AMC for the past six months, and there is interest in bringing another teacher to the
primary school if possible. At the last AMC meeting, the secretary presented a
summary of household demography, using data from the Community Self-Monitoring
books. The CSM books revealed what many people already suspected - that almost
half of the population in the AMC are under the age of 15, and that many of the girls
are not in school. However the CSM books also showed something that people hadn't
really realized, that households that are headed by a young female are often in the
lowest wealth ranking, and tend not to increase their asset base.
After discussing this problem during several meetings, the AMC decided that they
should look into ways to get more girls into school, look for opportunities for income
generation, and perhaps see if they can get a family planning officer to come to the
area to talk to them. When they heard from the local MAFF officer that some donor
organization was funding teachers for girl education, it seemed like a good
opportunity. The AMC has already agreed verbally that they can use some of the
money from the sale of surplus seed to build a house for an additional teacher, and the
Chief has given his support also and will help mobilize labour.
Mr. Miyanda has another appointment today also, to meet a local commercial farmer
about getting a new supply of maize seed for the following season. About every three
or four years, they try to organize to get some new seed, particularly for maize where
the preferred varieties are MMV400 and Pool 16. The last time they bought new
maize seed was three years ago, and the yields seem to have gone down since then,
although the methods they are using to improve soil fertility, such as green manure
LFSP Mid-Term Evaluation ARD
133
and inter-cropping, have helped a lot. They used to get new seed from CARE,
however now that the project stopped they have to buy it on the market. About 10 of
the AMCs in his part of the district have an agreement with this commercial seed
farmer to buy seed from him only, for a price which is cheaper than in the shops.
Money is going to be a little tight this year because the rains were not so good. So the
AMC is going to have to reduce the amount of new seed they can order from what
they originally wanted. However about half the farmers in the area had good sorghum
harvests this year, and they already have a buyer through the outgrower scheme. So in
a few weeks there should be some cash available which they can use to buy the new
maize seed. To cut down on transportation costs he wants to ask if they can use the
same trucks that are coming to pick up the crops to deliver some of the new maize
seed. Farmers that can't afford to buy the new stock of seed can still get seed from the
AMC seed bank on a loan basis.
Before he goes back to the village tomorrow morning, Mr. Miyanda also wants to pay
a visit to the District Agriculture Office. He wants to ask when they're going to
replace the extension officer for their area who left two months ago. The old officer
was quite active and worked closely with the AMC, but now he's afraid if they wait
too long to replace him, then people in the community will forget what they learned or
lose interest in extension. A couple of farmers in the AMC have also heard about a
kind of cover crop that improves soil fertility as well as conserves moisture, and has
asked if someone can come talk to them about this and maybe help set up a couple of
farm trials.
Finally, the District Education Officer calls Mr. Miyanda into the office. As they
shake hands, Mr. Miyanda thinks how strange this meeting would have been just 10
years ago. For a simple farmer to come meet the senior education officer of the
district was practically unheard of. However nowadays community organizations like
his are doing more things and getting more respect. Although he still feels a bit
nervous, he knows that he isn't just there by himself, he has the support of his entire
community, who over the last 15 years have learned to do what it takes to improve
their livelihoods and achieve food security.
LFSP Mid-Term Evaluation ARD
134
ANNEX IV - LSFP CONCEPTUAL FRAMEWORK
Livingstone Food Security Project Conceptual Framework
xxx = LFSP Intervention
LFSP Mission: To improve food security in drought prone areas in Southern Province
Food
Stocks
SO1
Rural Income
Food Aid
Income
Generating
Activities 7
Household
Assets
Food
Availability
Price
Agricultural
Production
Drought
Seed
Variety4
Water Harvesting &
Irrigation2
Labor
Land
Seed
Local Seed
Production 5
Seed
Quality
Commercial
Seed
Buyer
NRM Technology 3
Livestock
Production
Fertilizer
Post-Harvest
Management 6
LFSP Interventions
1 CBO mangement and capacity building empowers farmers to plan their own development and reduce
vulnerability to drought
2 Water harvesting structures constructed and rehabilitated
3 Conservation farming methods taught
4 Improved drought-resistant seed varieties introduced
5 Local seed banks established
6 Post-harvest management includes promotion of improved storage, food preservation, and processing
7 Small income generating activities developed based on agricultural produce and coping strategies
Capital
Food
Security
Quality
Volume
Marketing
Transactn Cost
Household
Size
Intra-Household
Dynamics
Liberalization
Education
Quality of
Life
Health
Forest/Soil
CBO Management 1
Coping
Strategies
LFSP Mid-Term Evaluation ARD
135
ANNEX V - TECHNOLOGY MATRIX
Purpose
The development of improved technologies (where the term "technology" is used in a
broad sense of the word including things like organizational structures, training
strategies, institutional linkages) is an important component of the LFSP. The
objective of this list is to summarize the main technologies that are being studied,
tested, or in used by the project.
The purpose of this exercise is to help the project communicate its technological focus
to other institutions, which may lead to dialog or replication on specific technical
aspects. It may also highlight areas where additional technical assistance is needed.
Columns in the matrix
Technology The new or improved method, technique or strategy being used in
the project or seriously considered
Stage of
Development
How far the development of the technology has progressed (e.g.,
under consideration, testing, demonstrations, application)
Extent of
Adoption
very limited - adopted by approximately 1-20% of beneficiaries
limited - adopted by 21-49% of beneficiaries
medium spread - adopted by 50-80% of beneficiaries
wide spread- adopted by 81-100% of beneficiaries
Remarks Whether the technology has succeeded, under what conditions
(e.g., which agroecological zones), if it failed why it failed, etc.
136
Technology Stage of
Development
Extent of
Adoption
Results/Remarks
1. Soil fertility improvement
Sunhemp testing limited, kalahari &
plateau
Demonstrations conducted over two seasons have shown good results but this was on limited scale. Possible
constraints to wider farmer adoption could arise from lack of implements to enable the incorporation of the
plants into the soil and to some extent seed availability as well as attitudes towards growing a crop which does
not seem to yield any direct benefit i.e non food. Impressive results are reported to have come from the plateau
zone on degraded soils.
Velvet bean testing limited, kalahari &
plateau
Demonstrations have given encouraging results but seed seems to be a problem to facilitate wider
demonstrations To encourage adoption training be given is highlighting the multiple use of velevet beans such as
soil moisture conservation and weed control
Tephrosia vogellia testing very limited,
kalahari & plateau
Seed was distributed during 1999/2000 season but farmers who got the seed could not plant due to unusual
rainfall pattern. It is expected that planting of the same will be done in the next season. The adaptability of this
plant species is therefore yet to be ascertained.
Sesbania sesbanb testing very limited,
kalahari & plateau
Farmers from LFSP went to Chipata to learn from fellow farmers on the plant management and benefits to crop
production. Four testing centres have been set up covering the plateau and kalahari zones where other farmers
are trained by fellow farmers on how to raise the plants. More time will be required to take the plants in the field
and demonstrate the benefits of improved fallow using Sesbania sesban on crop yields. So far there seems to be
no problems in terms of plant establishment and growth.
Gliricidia sepium testing very limited,
kalahari & plateau
A good candidate for testing in the Escarpment Zone in a more permanent cropping system, because land is a
greater constraint there than in either of the other two zones.
Felhidebia albida promotion wide spread LFSP involved in raising awareness on the importance of the species as a soil improver and the need for
selective cutting when preparing crop fields. A number of farmers are aware of the merits of this species which
occurs naturally.
Crop rotation promotion medium spread Through training LFSP has been encouraging farmers to adopt proper crop rotation sequences which involve
leguminous crops to enhance soil fertility maintenance and improvement. Benefits of this practice have been
widely appreciated in the project area.
Intercropping testing very limited Traditional mixed cropping practices are still common. Systematic intercropping practices are not likely to be
popular.
Improved fallows testing very limited
Shifting kraals promotion medium spread Awareness creation aimed at building on the farmers practices and making improvement in certain aspects such
as application methods to suit for instance conservation farming practices such as pot-holing. Need to continue
to test ways of combining agriculture and livestock in plateau.
2. Soil and Water Conservation
Contour
ploughing
promotion medium spread LFSP has given training to farmers on this technique. It has been widely adopted in the escarpment and kalahari
zones where it is been found appropriate for controlling gully erosion. For the benefits to be fully realised at
137
Technology Stage of
Development
Extent of
Adoption
Results/Remarks
individual farms cooperation from neighboring farmers is necessary. Such intervention should therefore target
the community as is the case under LFSP.
Pot holing testing very limited Pot-holing may not be appropriate in the Kalahari because the loose texture of the soils, but may prove effective
in the Plateau Zone where soils are heavier and can retain the moisture captured by the holes. It is time
consuming in the first year with the establishment of the holes but thereafter does not require large efforts of
maintenance. Limitations of the technology that might be affecting its adoption in the project area may include
increased labor demand and social stigma associated with using a hoe, a practice associated with low status due
to lack of cattle. Should be tested in the Plateau Zone. Farmer comments will be critical to understanding how to
adapt the technology to Plateau farmers conditions and constraints.
Ripping promotion very limited Affordability of the implement (Magoye Ripper) may be affecting the rate of adoption. The technology is also
said to be associated with increased labour for weeding arising from minimum tillage.
Stonelines promotion limited, escarpment This is constructed along contour to prevent silting of gabions.
Gabions testing medium spread,
escarpment
The technology has potential for wider adoption as it has proved to be effective for soil erosion control and
prevention of siltation in dams and streams where it has been used in the project area
Vetiver grass application medium spread Mainly used around water points and is a technology that has been widely appreciated and accepted. Vetiver
grass has been very effective where used in demonstrations on slopes in the Escarpment Zone. It is a labor
efficient technology with widespread application possibilities in all zones of the project where soil stabilization
and conservation are important to farmers. LFSP has plans to introduce the technology to crop fields where it
may be planted along contour ridges.
3. Water Harvesting and Sanitation
Dams/Weirs promotion medium spread,
plateau and
escarpment
Not suited to the Kalahari agroecological zone
Boreholes wide spread,
mostly kalahari
Appreciated by communities as the most appropriate way of accessing underground water. Communities have
responded positively in taking up the maintenance role.
Spring capping testing limited, escarpment Limited to areas with springs. Where springs have been constructed health problems associated with water
among humans are reported to have gone down
Indian Mark 2
pump
promotion wide spread Is not effective in pumping water from depths of more than 40metres
Bush pump Type
B
under
consideration
Can pump water from depths of up to 60 metres. This could be more suitable than the currently recommended
Indian Mark 2 in the project area where boreholes may need to go beyond 40 metres to find water. This pump is
found in Zimbabwe
Artesian well testing limited
Collector wells under Proposed option by consultant. May be useful in kalahari zone but ruled out because it is too expensive
138
Technology Stage of
Development
Extent of
Adoption
Results/Remarks
consideration
Solar pumping under
consideration
Initial cost could be high and unlikely to be afforded by the communities
Water shed
Management
testing medium spread Appreciated by communities who are now developing by laws to implement the desired practices
Water use zones promotion medium spread Water points zoned for separate human and animal areas
Pit latrines testing ver limited Didn't catch on
4. Irrigation
Farrow irrigation promotion Useful in conjunction with water lifting technology. Not effective in sandy soils
Treadle pump promotion limited Demonstrations have been on limited scale but results have been encouraging. There are plans to demonstrations
in Kalomo and Kasungula. Pumps have been given out on loan, but repayment has been challenging. After they
have been demonstrated in an area, LFSP stops providing loans and farmers have to buy them directly from the
manufacturers. To date no farmers have purchased treadle pumps, but several inquires have been made.
Drip Irrigation testing very limited on-farm testing of the technology being planned. This technology could be more adaptable to the kalahari
conditions as it provides higher water use efficiency.
5. Improved Crop Varieties (see also Annex VI)
Maize
MMV400 promotion wide spread Variety is preferred mainly on account of being early maturing and therefore provides early food during the
critical period. The small kernel size and hardness are however not liked by the farmers. Local varieties are still
planted.
Pool 16 promotion wide spread The variety seems to be more preferred over MMV400 because of having bigger kernels
Sorghum
Kuyuma promotion wide spread More adapted to the kalahari areas. Widely accepted because of earliness and being high yielding
Sima promotion limited Not as wide spread because it is not well adapted. It is a longer maturing variety compared to Kuyuma and as
requires a longer rainseason. Farmers however like it for its grain which is said to yield more and whiter flour.
Pearl millet
Kaufela promotion limited Not as well adapted as Lubasi. Seed distribution is limited to the kalahari areas
Lubasi promotion medium spread More adapted than Kaufela
Cowpea
Lutembwe promotion wide spread Improved varieties so popular that some farmers are likely to abandon the local varieties completely. More seed
of this variety is said to have been distributed compared to Bubebe
Bubebe promotion wide spread As for Lutembwe the variety has been successfully adopted mainly on account of its earliness
Groundnut
139
Technology Stage of
Development
Extent of
Adoption
Results/Remarks
Natal common promotion widespread well adapted and accepted
Folcon promotion limited Seed availability is a problem. Likely to have come from Zimbabwe
Chipego promotion limited adoption being affected by poor seed availability
Sweet potato
Chingovwa promotion widespread LFSP cannot meet demand for planting material
Zambezi promotion widespread Planting material is brought from outside the project area
Cassava
Bangweulu promotion limited The adoption of the use of cassava as a main food by the Tonga people, who currently use it only as a snack
could lead to increased adoption and production
Farming practices
Seed
multiplication
applied widespread See 10.2 page 87
Integrated pest
management
testing very limited
6. Livestock Health
Dipping schemes under
consideration
One of the activities planned for implementation under the livestock disease control program. A baseline survey
was being undertaken at the time of the evaluation team’s visit
Rotational grazing under
consideration
As for dipping schemes
Animal health kits under
consideration
As for dipping schemes
Donkeys for
draught power
testing limited Donkeys have been introduced in some parts of the project area though not through LFSP
Cattle vaccination under
consideration
One of the activities planned for implementation under the livestock disease control program. A baseline survey
was being undertaken at the time of the evaluation team’s visit.
Community
Livestock
Auxiliaries
promotion wide spread These have been identified and given training in livestock health management during 1998/99 season
7. Natural Resources Management
Controlled
burning
under
consideration
Some awareness creation through training has been undertaken. More needs to be done especially in making
traditional leaders work hand in hand with CBOs.
Selective cutting
of timber
under
consideration
One of the activities planned under the proposed joint forest management project in chief Sekute’s area. The
activity will also embrace the need to enable communities adequately benefit from the commercial exploitation
of these resources
140
Technology Stage of
Development
Extent of
Adoption
Results/Remarks
Selective cutting
for charcoal
under
consideration
limited One of the activities planned under the proposed joint forest management project.
Sustainable
harvesting of
Mopane
planned limited One of the activities planned under the proposed joint forest management project . Mopane poles are mainly
used in the construction sector and the trees are mostly found in the kalahari areas
Sustainable
harvesting of
Masuku
under
consideration
limited One of the activities planned under the proposed joint forest management project. Masuku trees are mostly
found on the plateau in Kalomo district
Sustainable
harvesting of
mawi fruits
under
consideration
limited One of the activities planned under the proposed joint forest management project. LFSP intends to increase the
value of this indigenous fruit by diversifying products through processing and help identify profitable markets.
Neem testing medium spread People in the project area appreciate the medicinal value of the Neem for humans and its pesticidal use in both
livestock and crops. There is a big demand for seedlings
Live fencingc under
consideration
One of the plant species earmarked for testing is sisal. This will be mainly for fencing gardens and likely to be
more appropriate around vegetable gardens especially in the escarpment area
Joint forest
management
testing limited some activities have been initiated in chief Sekute’s area. LFSP is playing the role of facilitator while the council
is coordinating the initiative and involving other stakeholders
8. Income Generating Activity
Curio marketing testing
vegetable
production
promotion medium The project is assisting the communities to consolidate the marketing arrangements through training in
understanding contractual obligations. Vegetable growers mainly in escarpment areas are being linked to traders
through development and dissemination of the market information.
beer brewing promotion widespread Project assisting women in improving the quality of beer brewed locally in order to increase profitability of this
enterprise
milk chilling
centers
planned Will increase amount of milk that can be marketed to Finta. Only one or chilling centers can be established
because of cost, so only areas near road will be served.
mat marketing consideration
9. Food Processing
Sorghum
dehulling
under
consideration
The promotion will focus in the kalahari zone and aim at creating market for the produce and hence encourage
production. Dehulling makes the sorghum white and improves marketability. Machines are expensive so will be
based at hammer mills.
Cowpea sausage testing limited Demonstrations done were successful and people in these areas are able make the product.
Yenga press testing limited Demonstrations for oil extraction from the Mungongo and other nuts. It holds some promise.
Market testing limited Proven effective in setting up a trading network for vegetable producers
141
Technology Stage of
Development
Extent of
Adoption
Results/Remarks
information
system
10. Monitoring and Evaluation
Community self￾monitoring
ledgers
promotion widespread Effective way to collect data on household level indicators. However proving difficult to ensure uniformity and
consistence in data collection among the communities
Food Production
Trends Survey
applied limited The data required is being collected by project staff. Sample size is rather small although representative. Data
processing and sampling has been a constraint.
Erosion pins Useful for participatory monitoring of erosion.
Topical appraisals promotion wide spread Used as a tool for helping to develop action plans to solve specific development issues
11. CBO Capacity Building
PRAs promotion wide spread Proven technique for getting information about communities and increasing participation. Variety of data
collection methods used. A series of PRAs were used to facilitate community involvement in identifying
development concerns and determining solutions before implementation of interventions.
Personal
empowerment
training
promotion wide spread Business oriented training.
Notes:
a. Tephrosia vogelli is a leguminous tree which produces nitrogen-fixing nodules in soils where demonstrations are being conducted. The
species is found wild in some parts of Zambia. In addition to use in soil improvement and as a tree crop (minimum 3 year cycle) the tree
produces the chemical tephrosium, which is a powerful insecticide. Tephrosium is also extremely toxic to mammals and cold-blooded
aquatic vertebrates (fish). Farmers in Southern Province (Kazungula and Kalomo, particularly), producing in an area without commercial or
public infrastructure for the provision of agricultural inputs, have an extremely great need for access to insecticides, both for soil and
vegetative (above ground) destructive crop insects. The tree has great potential for commercial uses in this regard. ICRAF has studied the
growth habits and chemical properties of the tree. LFSP may need to conduct experimentation to provide information on the use of
tephrosium and precision in processing, packaging and using the chemical for crop protection. When used for soil improvement, LFSP
could generate information on any residual toxicity in the soil following tree removal, as well as its effect on soil micro-organisms.
b. Sesbania has been used in widespread rows in fields with Escarpment type slopes (10-60%) in order to plant annual crops in between the
rows and thereby obtain field crop permanency in these areas with reduced or no fertilizer added to the annual crops. The tree rows have to
142
be culturally maintained with severe pruning so that the trees do not inhibit crop development. As an introduced leguminous tree species
roots of young plants should be checked for nodulation. If absent or only a few nodules are found, an appropriate rhizobium may be applied
to assure maximum nitrogen-fixation of the species in Southern Province soils. This tree would be good to test in the Escarpment zone.
c. Commiphora africana (mubwabwa) is used by farmers for live-fencing and it is found in the wild state within the project zone. Farmers
know how to use it for fencing which already enhances its possibilities for adoption. It appears to be very effective control for containing
livestock where needed. It is presently being used as fencing around vegetable gardens, especially in Escarpment Zone, and around
boreholes in Kalahari Zone. It requires approximately one year for establishment as an effective live-fencing technology in most areas.
LFSP Mid-Term Evaluation ARD
143
ANNEX VI - IMPROVED SEED VARIETIES CHARACTERISTICS
AND SOURCES5
Crop/Variety Characteristics Seed Source Breeding
Program
Maize
MMV400 110 days6
, O-P7
, short 1.4
to1.8 m stature, suitable for
drought prone areas, resistant
to streak virus, cob rot yield
potential 2.5/ha, white grain
ZamSeed Zambia National
Research,
Cereals Breeding
Program
GV412 Hybrid , early maturing Zamseed. Same as above
Pool 16 90 days, similar to MMV400
with excellent resistance to
strek virus.
Mpongwe
development
project (under
EEC Project)
Zambia National
Research,
Cereals Breeding
Program
Sorghum
Kuyuma 110 days, dwarf type 1.7m,
excellent resistance to
drought and disease,
potential yield 4-6 tons/ha,
white grain
Zamseed Zambia National
Research,
Cereals Breeding
Program
Sima 130 days to maturity
photosensitive, white grain,
does better on heavier soils,
6 tonnes/Ha, resistant to
disease, up to 2.3m high.
Zamseed Zambia National
Research,
Cereals Breeding
Program
Pearl Millet
Kaufela 90 days, O-P, short statured,
small grains, disease
resistant.
Western Province
Cereals Breeding
Program
(ZNAR)8
Zambia National
Research
Lubasi 90 days, O-P, short statured,
larger grain.
Western Province
Cereals Breeding
Program (ZNAR)
Zambia National
Agricultural
Research
Groundnut
Natal
Common
90 days, small kernel Zimbabwe,
ZamSeed
Legume
Breeding
Program
Chipego 110-120 days, yield 1.5 T/ha,
suitable for low rainfall
Food legumes
project under
Zambia National
Agricultural
 
5
 Includes seed, seedlings and cuttings depending upon the propagation material of the particular crop.
6
 To maturity, from sowing to harvest
7
 Open-pollinated
8
 Zambia National Agricultural Research
LFSP Mid-Term Evaluation ARD
144
Crop/Variety Characteristics Seed Source Breeding
Program
areas, larger kernel than natal
common, tolerant to leaf spot
disease.
FAO. Research,
Eastern Province
Legume
Breeding
Program
Falcon 90 days, other characteristics
similar to natal common.
SEEDCO Zimbabwe
National
Agricultural
Research .
Cowpea
Lutemmbwe Average yield of 1 ton/ha,
long pods with whitish cream
seeds, good leaf quality,
early (70 days) duration,
tolerant to cowpea aphid
born mosaic virus, resistant
to scab.
Legume Breeding
Program, ZNAR,
ZamSeed
Zambia National
Agricultural
Research
Bubebe Average yield of 1 ton/ha,
suitable for region 1 and 2,
early maturing (70days),
tolerant to the major insect
pests
Legume Breeding
Program, ZNAR,
ZamSeed
Zambia National
Agricultural
Research
Green gram
RSA 1 Yield 1-2 ton/ha, drought
resistant recommended for
region 1, enriches soil
fertility, short duration (75 to
80 days) maturity.
Food legumes,
FAO project.
Source from on
farm tests.
Bambara nut
Red Buddy 90-100 days, red
background. Not enough
information.
Farmer in Eastern
Province
Not enough
information
Brown Breed Days to maturity. Farmer in Eastern
Province
As above
Freckle red speckles, other Farmer in Eastern
Province
As Above
Sunflower
G100 Hybrid, Short stature, one
large head per plant, high oil
content
Zamseed National
Research
Record 90-100 days, soft shelled, has
high oil content bigger heads
than , drought tolerance.
SEEDCO,
Zamseed
As above,
Company of
Zimbabwe,
Annual Legumes for Soil Improvement
LFSP Mid-Term Evaluation ARD
145
Crop/Variety Characteristics Seed Source Breeding
Program
Sunhemp9 90 days to 50% flowering,
increase maize yields by
70% following sunhemp10
Magoye Family
Farm, GART
Soil
improvement
Program
Velvet bean Increases yield by70-80%. Magoye Family
Farm
Where did
Magoye Farm
obtain seed,
GART.
Cassava
Bangweulu 2 seasons to maturity, yields
over 31.27 t/ha, resistant to
cassava mosic virus,
moderate resistance to the
cassava mealy bug.
Mr. Chitundu,
ZNAR, Luapula
Province
Root and Tuber
Breeding
Program, ZNAR
Sweet potato
Chingovwa Most preferred and most
grown in Zambia, erect type,
deep lobbed leaf shape, high
dry matter content, over 18
tones per ha. Poor storability.
Mr. Chitundu,
ZNAR, Luapula
Province.
Root and Tuber
Breeding
Program, ZNAR
Zambezi Deep orange (Carrot0 flesh
color associated with high
carotene content useful for
sight, slightly more waterly
than Chingovwa.
Mr. Chitundu,
ZNAR, Luapula
Province.
Root and Tuber
Breeding
Program, ZNAR
Trees and Shrubs
Mubwabwa
(Commiphora
africana)
Live-fencing used around the
vegetable gardens, boreholes
and Kraals in some areas of
the project area, propagated
by cuttings. 3-5months
fencing establishment (?)
Drought resistant. Currently
there about 6 group gardens
using the technology and
about 5 boreholes.
Local indigenous
sources
Cedrella Used for medicinal purposes Exotic to the
project area.
Neem
(Azadirachta
indica)
Used for cattle dips,
medicinal purposes, Pest
control, shade
Magoye Family
Farm
Integrated pest
management.
Guava Not available Private and
MAFF nurseries.
 
.
10 Empirical evidence. Do we have any hard data on N-fixation amounts, C/N ratios, or organic matter
accumulation in soils following sunhemp in the rotation? NO
LFSP Mid-Term Evaluation ARD
146
Crop/Variety Characteristics Seed Source Breeding
Program
Pawpaw As above As above
Mulberry As above As above
Mango As above As above
Nsombo /
Water Berry
(Syzgium
cordatum)
Local fruit abundant in the
project area at certain times
of the year, project trying to
figure out how best it can be
processed.
Sesbania
sesban
Used for soil fertility
improvement in LFSP on￾farm demonstrations
ICRAF Soil fertility
improvement
program
Tephrosia
vogelii
Used in LFSP soil
improvement on-farm
demonstrations, can fix
nitrogen, good tolerance to
acid soils11
ICRAF Soil fertility
improvement
Program, Eastern
Province
 
11 Tephrosia has a very high commercial potential for use as biological-based agricultural chemical for
control of several crop damaging insects. Potential for effective cattle dip for tick control. Controlled
laboratory and field testing required. High toxicity to goats and fish.
LFSP Mid-Term Evaluation ARD
147
ANNEX VII - SCOPE OF WORK
Concurrent Evaluation of Three of USAID/Zambia Activities:
1) Cooperative League of the USA Rural Group Business Program
(CLUSA/RGBP);
2) CARE Livingstone Food Security Project (CARE/LFSP);
3) Wildlife Conservation Society’s Administrative Management
Design Project (WCS/ADMADE)
STATEMENT OF WORK
Article 1. Introduction
With regard to the three projects identified in the title of this statement of work,
USAID/Zambia would like to find out whether investments in profit oriented farmer
group businesses (CLUSA), food security oriented village management committees
(CARE), and wildlife conservation oriented village action groups (WCS) have had or
are having a beneficial impact. If so, USAID/Zambia would like to identify the
elements of successful investments that can be replicated to improve ongoing or
future investments. Finally, if an investment were not achieving the intended results,
USAID/Zambia would like to know how to reorient that investment so that it does
achieve the intended results.
In support of Zambian economic liberalisation, USAID/Zambia has initiated and
supported activities that stimulate rural economic growth since 1991. Under
USAID/Zambia's Country Strategic Plan for the 1998 - 2002 period, Strategic
Objective 1 (SO 1) is "increased rural incomes of selected groups." Approximately 6
million of Zambia’s 10 million people live and work in rural areas.
SO 1 investments aim at increasing the incomes of rural families working together as
farmer group businesses, village management committees or village action groups.
Hopefully, rural families working as groups will result in more cost effective (and less
risky) technology dissemination, training, rural finance, output marketing and wildlife
management service delivery. Lower service delivery costs will contribute to more
sustainable, customer responsive and profitable service delivery agencies. Finally,
more sustainable and profitable service delivery will result in increased rural family
opportunities to improve their productivity and incomes.
USAID/Zambia recognizes the importance of Zambia’s macroeconomic and sectoral
policy environment. Investments that focus on reducing service delivery costs and
raising rural family productivity are likely to identify and lead to the resolution of
“second generation” policy constraints. USAID/Zambia investments ground truth
neo-classical economic theory based predictions about market driven resource
allocation and use and hopefully generate ideas on how public and private institutions
can best contribute to improved rural family welfare. USAID/Zambia regards its
service provision investments as applied research.
LFSP Mid-Term Evaluation ARD
148
Actual SO1 activities spring from rural family problem and opportunity identification.
They are intended to encourage rural family contributions to solving their social or
economic problems, enhance women's contribution to rural economic growth and
encourage government food security and rural finance policies that promote private
initiative.
During the April – May 2000 period three of SO1’s projects will be evaluated.
CLUSA/RGBP and CARE/LFSP are earmarked for mid-term evaluations while the
WCS/ADMADE evaluation will be an End of Project Evaluation.
As the result of an unsolicited proposal from CLUSA, the Rural Group Business
Project began in May 1996. This 5 year, $5 million activity promotes the emergence
of democratically self-managed, financially viable group businesses that improve
rural family incomes. Since its inception CLUSA-RGBP has modified its group
business development approach. It now focuses specifically on small farmer high
value crop production usually under forward contract to agro-processors. CLUSA￾RGBP credit provision is almost entirely for seed and fertilizer.
Another unsolicited proposal, this time submitted by CARE International, resulted in
the Livingstone Food Security Project. This 5 year $3.6 million project began in July
1996. The project promotes community institution management of drought resistant
crop seed multiplication and distribution, soil conservation, water harvesting,
marketing, and some income generating activities. As a result of CARE’s activities
rural family food stocks have increased in some of Zambia’s most drought prone
areas.
The third project to be evaluated, as an end of project evaluation, adds a bit of
complexity to this activity. Since 1989 USAID has supported Zambia’s
Administrative Management Design (ADMADE) Project and the National Parks and
Wildlife Service with funding made available through the Regional Natural Resources
Management Project. Funds were initially managed by USAID’s regional office in
Harare but eventually project management was vested in USAID/Zambia with
funding obligated through bilateral project agreements. Over the 10 years of project
life, implementation vehicles included a grant to the World Wildlife Fund, funds
made available directly to the National Parks and Wildlife Service through Project
Implementation Letters, short-term technical assistance in Wildlife Conservation
Revolving Fund capacity building and, finally, since October 1998, a Cooperative
Agreement with the Wildlife Conservation Society of New York as the result of an
unsolicited proposal. The WCS activity, entitled the ADMADE Sustainability Project,
was a 15 month, $.461 million activity that ended on December 31, 1999.
The overall 10 year RNRMP/ADMADE investment sought to introduce and develop
the idea of community wildlife management in Zambia, including use of village
wildlife scouts and the sharing of hunting revenues with protected area communities
for their use in improving their livelihoods. Community involvement in wildlife
management is now a stated national policy although the Zambian government’s
wildlife institutions are currently in a state of significant transition. The WCS
ADMADE Sustainability Cooperative Agreement was intended to document
LFSP Mid-Term Evaluation ARD
149
ADMADE lessons learned and research findings hopefully to inform future USAID,
other donor and GRZ investments in wildlife management.
Article 2. Overall Orientation of the Consultancy
The consultancy will comprehensively assess the three projects. USAID/Zambia
would like each project evaluation to result in a separate evaluation report. However,
by evaluating the three activities under one contract USAID seeks lessons learned that
may be applicable to all three project objectives (rural incomes, food security, wildlife
management) in order to positively influence ongoing or future activities or
investments. Therefore, a fourth report encapsulating lessons learned and describing
their implications across activity objectives is required.
To the greatest extent possible USAID would like the evaluations to provide
quantitative evidence of investment impact on rural incomes (CLUSA), food security
(CARE) or wildlife management (RNRMP/ADMADE). Quantitative evidence should
be presented over time to illustrate any growth or reduction in investment impact
during project implementation. Where quantitative evidence is not available or
relevant, qualitative descriptions of impacts and processes will be required.
With regard to CLUSA RGBP and CARE LFSP, the consultancy should assess
project impact and identify ways to improve implementation, if necessary. The
consultancy should recommend whether USAID/Zambia should consider extending,
expanding or cutting short the projects. Finally, the consultancy should package
relevant findings so that systemic or national level impact from evaluation lessons
learned might be achieved with specific reference to the Zambian context.
The RNRMP/ADMADE evaluation in many ways is a traditional end of project
evaluation. However, as laid out in the recent “Final Report: Assessment of
Community Based Natural Resource Management (CBNRM) in Southern Africa”
(August, 1998) ADMADE represents an opportunity for comparing the Zambian
community wildlife management experience with other wildlife management lessons
learned under RNRMP and throughout the world. The last 15 months of
RNRMP/ADMADE has resulted in substantial empirically based information on the
impact of ADMADE on communities and wildlife in 9 of Zambia’s 34 Game
Management Areas. Finally, the CARE and CLUSA experiences may have something
to say about how community capacity to manage natural resources, and the benefits
accruing from natural resources management, can be increased. Again, the
consultancy should package relevant findings so that systemic or national level impact
from evaluation lessons learned might be achieved with specific reference to the
Zambian context.
An external team, with appropriate local participation, will conduct the evaluation of
the three projects. The team is required to respond, in concisely written reports, to all
points and questions included in the scope of work.
Article 3. Proposals, Evaluation Criteria
LFSP Mid-Term Evaluation ARD
150
USAID/Zambia would like to use the Raising Agricultural Incomes in a Sustainable
Environment (RAISE) Tier 3 process in awarding this contract. Contractors are
required to submit their technical proposals (i.e. without costs) to USAID/Zambia.
The proposals should include a draft version of the contractor’s workplan,
methodology and suggested personnel for conducting the assessment. The technical
proposals will be graded according to the following criteria:
Methodology: Ability to: a) identify results desired under the project and
generate quantitative indicators of project impact where possible and qualitative
indicators where quantitative indicators are not possible; b) identify beneficiary
perceptions of project delivered services and beneficiary participation in the project;
c) generate information on partner or stakeholder perceptions of the projects; d)
generate lessons learned across projects in line with scope of work questions; e)
present findings in a use friendly and compelling manner.
Total Points: 50 points out of 100
Personnel: Appropriate professional training at the Masters of Science level
or above, experience in evaluating USAID projects in agribusiness, food security,
natural resources management or community mobilization, experience writing
technical documents based on the compilation of field visit findings, experience in
presenting evaluation findings in a user friendly and compelling manner, experience
in Africa and experience in Zambia.
Total Points: 30 points out of 100
Draft Workplan: Ability to deliver a highly competent team to arrive and
work in Zambia, all at the same time, over a period of five 6-day work weeks, conduct
the evaluation in a way that comprehensively answers Scope of Work questions, and
deliver the required deliverables by COB, March 3, 2000.
Total Points: 20 out of 100
 Following receipt of proposals, USAID will review the documents and select a
suitable offeror. Technical proposals should be sent to:
David Soroko
SO1 Team Leader
USAID/Zambia
351 Independence Avenue
Lusaka, Zambia
Fax: 1- 254532
E-mail: dasoroko@usaid.gov
Cost proposals should be sent to:
Beatrice Lumande
USAID/RCSA
Plot 14818 Lebatlane RD
Gaborone West, Ext 6
LFSP Mid-Term Evaluation ARD
151
Gaborone
Botswana
Fax: 267324486
E-mail: blumande@usaid.gov
End date for receiving both technical and cost proposals is March 3, 2000 at 12.00
noon.
Article 4. Scope of Work
Following is the scope of work for each project.
4.1 CLUSA RURAL GROUP BUSINESS PROGRAM MID TERM
EVALUATION
4.1.1 Background
The five year, $5 million Cooperative League of the USA (CLUSA) Rural Group
Business Program (RGBP) began in May 1996. The project, currently working in
four districts of Zambia (Mumbwa, Chibombo, Mazabuka and Monze), was aimed at
promoting the emergence of democratically self-managed, financially viable group
businesses that improve rural family incomes. Using fully costed credit for rural
groups, CLUSA brought to Zambia its rural group development experience gained
worldwide including West Africa. The Cooperative Agreement with USAID
indicated that in five years 210 rural groups with a total membership of 9,450 farmers
would have been participating in the program. During the five years of project
implementation, cumulative credit of $5 million would be disbursed to the groups
whose membership would be 30% women. Also, at the end of five years, it was
expected that 80% of the group businesses would have good managerial skills, access
to in-house finance through accumulated profits, and regular and dependable access to
inputs and markets.
4.1.2 Evaluation Objective
The primary CLUSA/RGBP evaluation objective is to determine whether USAID
investments are achieving their desired impact, why or why not. A second objective is
to generate ideas on how the impact of USAID investments in CLUSA/RGBP
activities can be improved. A final objectives is to generate ideas on how
CLUSA/RGBP experiences can influence ongoing or future USAID and other
institution investments in increasing rural incomes, improving food security, and
managing natural resources.
4.1.3 Evaluation Questions
1. What are the results identified in the cooperative agreement? Who are the
beneficiaries? Have CLUSA/RGBP activities to date made progress in achieving
those results? Why or why not? Present your findings with regard to annual results
LFSP Mid-Term Evaluation ARD
152
and impact quantitatively and using graphs where appropriate. Has the program
made significant contributions to USAID’s “increased rural incomes of selected
groups” Strategic Objective in line with the SO’s results framework?
2. How is the project implemented? What are the most important components of
project implementation? How was the project’s location identified? How much
project financing is expended in Zambia (actual and percentage figures)? What
percentage is expended in Lusaka and what percentage is expended in rural areas
where CLUSA works?
3. Is the project demand driven? Do beneficiaries find it relevant to their
circumstances? How does the project identify what the beneficiaries want? Is this
approach effective in identifying what the beneficiaries want?
4. What are the most important services the project delivers to rural families?
How
were these services identified? How are they delivered? Are they delivered cost
effectively? Is their delivery effective in Zambia’s rural context? Could other
institutions deliver these services if CLUSA did not? Could other institutions
deliver CLUSA like services if they so desired? In terms of incentives, finance,
personnel resources and other variables what would other institutions need to
deliver similar services? Has CLUSA worked with local institutions to foster
continuation and sustainability of programs and services when the project ends?
5. Is there significant participation by women in the rural group business
program?
Is the program beneficial to women participants? Why? How can more women
participate in and benefit from the program?
6. What are the social and economic characteristics and organization of project
supported group businesses? What are their relative strengths and weaknesses
with regard to business capacity, income and investment management, relations
with agribusiness, knowledge and utilization of agricultural technologies, and skill
levels to undertake additional welfare enhancing activities? What additional skills
may be required to make rural group businesses effective and self-reliant beyond
USAID assistance?
7. Is the program well organized to allow for cost effective implementation?
Does it
require any significant structural changes? Does the program offer opportunity for
the establishment of sustainable group businesses development service delivery
agencies beyond USAID assistance? Should it?
8. What partnerships with other public or private sector agencies has
CLUSA/RGBP
made that enhance project service delivery and impact? What partnerships might
CLUSA/RGBP make that would improve service delivery and impact?
LFSP Mid-Term Evaluation ARD
153
9. What has Credit Management Services contributed to CLUSA/RGBP project
implementation? What are the strengths and weaknesses of CLUSA/RGBP’s
partnership with CMS for credit management?
10. Are there any significant policy constraints to program implementation? Is the
program supportive of the stated Zambian government policy of agricultural
liberalization and establishment of a private sector led economy? Has government
policy influenced the program? How? Has the program influenced government
policy? Why or why not?
11. What lessons learned during project implementation could lead to improved
CLUSA/RGBP impact? What lessons learned should inform decisions on project
time and finance extension or expansion?
12. What lessons learned during project implementation might influence ongoing
or
future USAID investments in food security, rural incomes or natural resource
conservation?
13. What are the advantages and disadvantages, particularly to beneficiaries and
USAID, of extending, expanding or cutting short the CLUSA/RGBP Cooperative
Agreement?
14. Given the responses to the above questions, how can USAID/Zambia best
utilize
lessons learned from the implementation of this activity to inform government
policy dialogue and future government, donor or private sector investments?
4.1.4 Performance Reports and Previous Project Assessments
As required in the Cooperative Agreement, CLUSA prepares quarterly and annual
performance reports that are submitted to USAID/Zambia. Prior to the start of every
new activity year, the project staff submits an annual workplan. CLUSA also have a
length of project monitoring plan in place.
Two internal assessments of the rural group business program were undertaken in
1999. The first assessment focused on CLUSA/RGBP technology dissemination
activities. It was undertaken in May – June and is entitled “Less Hunger, More
Money, CLUSA: Making a Difference in Zambia.” The second assessment was an
internal CLUSA assessment and was entitled “Internal Assessment of the Zambia
Rural Group Business Program (RGBP).” It was undertaken in July – August, 1999.
CLUSA/RGBP, CARE/LFSP and ADMADE impact monitoring system were
described in a document entitled “A Profile of Community Based Monitoring Systems
of Three Rural Development Projects in Zambia” in November, 1998. In addition, the
CLUSA program coordinator has made two written presentations, in Nairobi and
Washington respectively, of the program. These and other related reports will be
made available to the selected contractor at the start of contract implementation.
LFSP Mid-Term Evaluation ARD
154
4.2 CARE LIVINGSTONE FOOD SECURITY PROGRAM MID-TERM
EVALUATION
4.2.1 Background
CARE Livingstone Food Security Project (CARE/LFSP) started as the South West
Drought Relief program in October 1994, and obtained USAID funding in July 1996
to address fundamental causes of food insecurity in Kalomo, Livingstone, and
Kazungula districts of Southern Province. LFSP is a five year $3.6 million project.
Four mutually re-enforcing objectives were established:
- Community and institution capacity building;
- Improved and sustainable farming systems;
- Water harvesting and utilization;
- Increased incomes and income-earning opportunities.
Under Community and Institution Capacity Building CARE/LFSP was to assist
18,000 farmers organized into village management committees within three years. For
the development of improved and sustainable farming systems CARE/LFSP would
introduce and facilitate distribution of a diverse range of drought tolerant seed to
improve productivity and raise participating farmer incomes. CARE/LFSP would also
assist rural families by introducing soil moisture conservation and management
practices and techniques to increase soil fertilizer and water harvesting. Finally,
CARE/LFSP planed on increasing the incomes and income earning opportunities of
participating families through expansion of trading and marketing.
4.2.2 Evaluation Objectives
The primary CARE/LFSP evaluation objective is to determine whether USAID
investments are achieving their desired impact, why or why not. A second objective is
to generate ideas on how the impact of USAID investments in CARE/LFSP activities
can be improved. A final objective is to generate ideas on how CARE/LFSP
experiences can influence ongoing or future USAID and other institution investments
in increasing rural incomes, improving food security or managing natural resources.
4.2.3 Evaluation Questions
1. What are the results identified in the Cooperative Agreement? Who are the
beneficiaries? Has CARE/LFSP made progress in achieving those results? Why or
why not? Present your findings on an annual and overall basis. Has the program been
successful in making significant contributions to USAID/Zambia’s SO 1 in line with
the results framework?
2. How is the project organized and implemented? What are the most important
components of project implementation? How was the project’s location identified?
How much cooperative agreement financing is expended in Zambia (actual and
percentage figures)? What percentage is expended in Lusaka and what percentage is
expended in rural areas where CARE/LFSP works?
LFSP Mid-Term Evaluation ARD
155
3. Is the project demand driven? Do beneficiaries find it relevant to their
circumstances? How does the project identify what the beneficiaries want? Is this
approach effective in identifying what the beneficiaries want? How effectively do the
beneficiaries participate in project implementation?
4. What specific services does the project deliver to rural families? How are these
services identified? How are they delivered? Are these services delivered cost￾effectively? Are the services relevant to rural families? Could other institutions
deliver these services if CARE/LFSP did not? In terms of incentives, finance,
personnel resources and other variables what would other institutions need to deliver
similar services? Has CARE worked with local institutions to foster continuation and
sustainability of programs and services when the project ends?
5. What partnerships with public or private sector institutions has the project created
to enhance the delivery of services to rural families? What additional partnerships
might enhance service delivery?
6. Is there significant participation by women in the project? Is the program
beneficial to women participants? Why? How can more women participate in and
benefit from the project?
7. What are the social and economic characteristics and organization of project
supported village management and area management committees? What are their
relative strengths and weaknesses with regard to capacity building, income and
investment management, linkages with agribusiness, knowledge and utilization of
agricultural technologies, and skill levels to undertake additional welfare enhancing
activities? What additional skills may be required to make these institutions more
effective and self -reliant especially beyond USAID assistance?
8. Are there any significant policy constraints to program implementation? Is the
program supportive of stated Zambian government policy of agricultural liberalization
and establishment of a private sector led economy? Has the project been influenced
by government policy? Why or why not? Has the project influenced government
policy? How?
9. What lessons learned during CARE/LFSP implementation could lead to improved
CARE/LFSP impact? What lessons learned should inform decisions on potential
extensions to the project time frame? potential increases in project financing? What
are the advantages and disadvantages, particularly to beneficiaries and USAID, of
extending, expanding or cutting short the CARE/LFSP Cooperative Agreement?
10. What lessons learned from the CARE/LFSP implementation could lead to
improved future USAID investments in food security, rural incomes and natural
resource conservation?
11. How can USAID/Zambia best utilize the lessons learned to inform Zambian food
security, agricultural extension and natural resource management policy dialogue?
LFSP Mid-Term Evaluation ARD
156
4.2.4 Performance Reports and Previous Project Assessment
As required in the Cooperative Agreement, CARE prepares quarterly and annual
performance reports that are submitted to USAID/Zambia. Prior to the start of every
new activity year, the project staff submits an annual workplan. A monitoring and
evaluation plan for the entire cooperative agreement time period is in place.
“End of Phase I Report” was produced in June 1996. A “Marketing Consultancy,”
which came out more like a project evaluation, was completed by the Participatory
Assessment Group in November, 1997. A “Seed Scheme Assessment: (1994-1998)
was completed by CARE’s Monitoring, Evaluation and Research Unit in November,
1998. A “Marketing Study” for CARE/LSP was carried out in December 1998. A
USAID intern wrote “A Review of Monitoring in the Livingstone Food Security
Project: Trip Report” in September, 1998. CLUSA/RGBP, CARE/LFSP and
ADMADE impact monitoring systems were described in a document entitled “A
Profile of Community Based Monitoring Systems of Three Rural Development
Projects in Zambia” in November, 1998. CARE/LFSP conducted an internal mid term
review titled “Work Ends, Knowledge Endures: Lessons for the Process for
Extension, Expansion and Replication” in June – July 1999. The reports will be made
available to the selected contractor at the start of contract implementation.
4.3 ADMADE END OF PROJECT EVALUATION, SCOPE OF WORK
4.3.1 Background
With Regional Natural Resources Management Project (RNRMP) financing
ADMADE was initiated in August 1989 as a community-based wildlife conservation
program in 9 of Zambia’s 34 Game Management Areas (GMAs). A total of $4.8
million has been invested in the project. It ended on December 31, 1999.
The Project Paper Supplement laid out the following project purposes:
- To increase involvement of local communities and private interests in
sustainable management and use of wildlife resources;
- To test the viability and replicability of community based natural resources
management and use, and integrate programs into existing NPWS services;
and,
- To demonstrate the effectiveness and legitimacy of community capacity
building in wildlife management as a profitable and sustainable land use
option in GMAs.
Over the years, the program evolved to include various community development
activities as well as diversification of income opportunities. In addition to USAID
regional and bilateral Missions, institutions involved in the management of the
RNRMP/ADMADE program were the Ministry of Tourism (policy direction) the
former Department of National Parks and Wildlife Services - NPWS (now the Zambia
LFSP Mid-Term Evaluation ARD
157
Wildlife Authority (ZAWA)) and within NPWS the Wildlife Conservation Revolving
Fund (WCRF). The Nayamaluma Institute provided research and training services for
Community Based Resource Management.
The Project Paper Supplement identifies program outputs as follows:
- Improvement of Ministry of Tourism policies related to private sector efforts
in conservation and tourism;
- Improvements to the operations of the Wildlife Conservation Revolving Fund;
- Assistance to land use planning; and,
- Training in managing wildlife resources.
Between 1989 and 1994 USAID provided NPWS with training, commodities and
technical assistance in establishing the ADMADE program. Between July 1994 and
December 1995 under a Cooperative Agreement, WWF Inc. provided NPWS with
technical assistance in the implementation of the ADMADE program (legislative
reform, participatory GMA planning and improvements to information systems).
Between July 1996 and July 1998 USAID provided ADMADE financing directly to
the National Parks and Wildlife Service through Project Implementation Letters.
Between October 1998 and December 1999, under a Cooperative Agreement, WCS
provided technical assistance to document and disseminate ADMADE lessons learned
and impact.
4.3.2 Evaluation Objectives
The primary RNRMP/ADMADE evaluation objective is to determine whether
USAID investments achieved their desired impact, why or why not. A second
objective is to generate ideas on how the impact of USAID investments in community
wildlife management might have been improved. A final objective is to generate ideas
on how RMRMP/ADMADE experiences can influence ongoing or future USAID and
other institution investments in natural resources conservation, increasing rural
incomes or improving food security
The selected consultant will do a brief synopsis of the findings of evaluation and other
documents between 1989 and 1995, and carry out an evaluation of the project’s
performance with reference to original project objectives and USAID's strategic
objectives between 1996 and 1999. This approach is intended to make the evaluation
more manageable and less reliant on interviewee recall for the years before 1996.
4.3.3 Evaluation Questions
1. What are the results identified in the project paper supplement and the WCS
cooperative agreement? Who are the beneficiaries? Were program goals, objectives,
outputs and beneficiaries clearly identified and understood by the implementing
agencies? Have ADMADE activities achieved those results? Why or why not?
2. Summarize the major findings of the various evaluations carried over the life of the
RNRMP/ADMADE project? What did the evaluations say about ADMADE’s ability
LFSP Mid-Term Evaluation ARD
158
to mobilize community contributions to wildlife management? What did they say
about ADMADE’s ability to influence national policy? about ADMADE’s ability to
deliver tangible economic or social benefits to rural communities? about ADMADE’s
ability to conserve wildlife and discourage illegal hunting? What did previous
evaluations say about the role of the Nyamaluma Training and Research Center in
ADMADE implementation?
3. How did the program management and institutional arrangements evolve over its
life span? Did this evolution have any positive or negative impact on the achievement
of RNRMP and ADMADE objectives? Focus this discussion on USAID and GRZ
project management and institutional arrangements as well as institutional
arrangements in the project areas.
4. Beginning the analysis in 1996, how was the project organized and implemented?
Was implementation effective? Did implementation focus resources on the most
important wildlife conservation and community development problems and
opportunities? What was the role of the Wildlife Conservation Revolving Fund in
ADMADE implementation? What was the role of the Nyamaluma Training and
Research Center?
5. Describe ADMADE relationships with the Ministry of Tourism, other public
institutions nationally and in the project area (relevant to project objectives), local or
“traditional” institutions (such as Chiefs and village headmen), private sector
operators and Game Management Area communities. Did these relationships
contribute to achievement of project or cooperative agreement objectives? Why or
why not? How effectively has the project collaborated with private interests in
tourism (GMA communities, tour operators, professional hunters, lodge or safari
camp owners)? Has ADMADE worked with local institutions to foster continuation
and sustainability of programs and services after the project ends? Has this been
successful in developing the capacity for local institutions to provide ADMADE
services now that USAID financing has ended?
6. Describe the nature and organization of community based institutions supported by
the project. How participatory are these institutions in terms of wildlife management
and investment decision making? Was there significant participation by women in the
program? Was the program beneficial to women? Why? How can more women
participate in and benefit in community wildlife management? What are the relative
strengths and weaknesses of women and men with regard to wildlife management,
revenue sharing and revenue reinvestment, and linkages with tour operators and
professional hunters?
7. What is the overall program impact on wildlife populations, household incomes,
rural family quality of life, community capacity building, and land use planning?
Please quantify and present graphically, on an aggregated and per capita basis,
investments in Game Management Areas (emanating from safari hunting, donors,
private investors, USAID, etc.) attributable to ADMADE and wildlife conservation.
LFSP Mid-Term Evaluation ARD
159
8. What income earning opportunities have community groups pursued? What
specific aspects of those activities make them attractive? What potential income
earning activities were not pursued by communities? Why not?
9. What has been the progress against each of the four program objectives? What
factors influenced results achievement? For which program objectives has progress
been more difficult? Why? What have been the major constraints to the achievement
of the program objectives and outputs? What have been the major factors contributing
to achievements?
10. What government policies or orientations have facilitated or hindered the
achievement of the program objectives? Has RNRMP/ADMADE influenced national
natural resources management policy? Why or why not? Has this influence been
important?
11. With regard to recent ADMADE food security initiatives, are there lessons GMA
communities can beneficially learn from CARE and CLUSA in the areas of seed
multiplication and distribution, income generation, business skills training, linkages
with agribusiness? Are CARE and CLUSA like activities appropriate for natural
resource conservation in GMAs? Do CARE and CLUSA offer approaches relevant to
Community Resource Board needs?
12. Has the program been successful in making significant contributions to
USAID/Zambia’s SO 1 in line with the results framework?
13. What lessons learned from RNRMP/ADMADE implementation and evaluation
are important for future USAID investments in food security, rural incomes and
natural resource conservation? What lessons learned can inform future donor, GRZ
and private sector investments in community wildlife management?
4.3.4 Performance Reports and Previous Project Assessments
Important and relevant reports include “The Reorganization and Restructuring of the
Department of National Parks and Wildlife Services (1992), “Report on Financial
Management of the Wildlife Conservation Revolving Fund” (1993), “NRMP –
Zambia Component of the Southern Africa Regional Project, A Success in the
Making” (1995) (which resulted in a Project Paper Supplement), “A Report to USAID
and Ministry of Tourism’s Department of National Parks and Wildlife Services on a
Suitable Community Based Wildlife Management Mechanism” (1995), “Report of the
WCRF Financial Management Capacity” (1998), “An Evaluation of the ADMADE
Program: With Special Reference to the Strengthening Phase” (1998), “Final Report:
Assessment of Community Based Natural Resource Management in Southern Africa
(August 1998), “A Profile of Community Based Monitoring Systems of Three Rural
Development Projects in Zambia” (November, 1998). Between October 1998 and
December 1999, several special studies papers were produced to document the
ADMADE process and results. The selected consultant will have access to these
reports.
LFSP Mid-Term Evaluation ARD
160
Article 5. Level of Effort, Team Composition and Timing, Logistical
Support
It is anticipated that the three person consultancy will be for 5 work weeks in April –
May 2000, with an additional and concurrent one person, two work week effort by an
evaluation packaging/desktop publishing expert at the end of the consultancy.
USAID/Zambia will use a fixed fee performance based contract as an instrument for
conducting this evaluation. Accordingly, although USAID/Zambia suggests that the
team be composed of an agricultural/agribusiness, food security/community
organization, natural resources/wildlife conservation specialists, with local
participation for additional Zambian specific expertise, and a two work week
contribution by an evaluation packaging/desktop publishing expert, it is incumbent
upon the contractor to determine the number of persons as well as their expertise for
USAID/Zambia’s consideration. It is essential that at least one of the core team
members has proven USAID project evaluation experience. With regard to Zambian
experts included in the team, contractors need to take due regard of prevailing USAID
local employment compensation levels.
5.1 Duty Post: The contractor shall perform all the work under this activity in
Zambia.
5.2 Logistical Support: The contractor is responsible for providing in-country
transportation and secretarial support while in Lusaka. The consultant will also make
own field trip travel arrangements. USAID/Zambia or local partners may be
consulted on logistics of sourcing field transport. It must be noted that
USAID/Zambia will not be able to provide any office space for this consultancy.
5.3 Work Week: A 6-day workweek is authorized.
Article 6. Reporting Requirements / Deliverables
6.1 Commencement
During the first week of the team’s presence in Zambia, the consultant’s will meet
with the SO1 team leader and his staff to answer questions, clarify tasks, obtain
relevant contacts, obtain documents and establish an implementation plan
6.2 Draft Report
After twenty (20) working days of contract implementation, the team will submit a
draft summary report to USAID (5 copies of each project). The draft report will
summarize major findings and recommendations. Three working days after this
submission, the team will make a presentation to USAID, the government of Zambia
and other select partners. The presentations will briefly describe the methodology and
summarize the preliminary findings, conclusions and recommendations of the
evaluations. The team will take note of the oral questions and comments from
meeting participants. The team will then have 7 working days to finalize the report.
LFSP Mid-Term Evaluation ARD
161
6.3 Final Report
After thirty (30) working days of contract implementation, the consultant shall deliver
the final report to USAID. The final report shall address all comments from the
review meeting in 6.2 above. Ten (10) hard copies of the evaluation report of each
program and an electronic copy in Word 97 must be submitted.
The final project evaluation reports shall be concisely written and include an
Attractive Cover Page, Table of Contents, Executive Summary, List of Acronyms, the
Main Report in compliance with the Scope of Work, a Statement of Conclusions and
a Statement of Recommendations. The body of each of the reports must describe the
relevant country context in which the project was developed and carried out, and
provide the information on which conclusions and recommendations are based. The
reports must present quantitative evidence of project impact whenever possible using
graphs and tables. Sidebars of success stories are also requested, where appropriate.
The reports must include attractive photographs of project activities either taken by
evaluation team staff or obtained from USAID/Zambia. The final report must be as
user friendly as possible. Depending on the findings, the reports may provide the basis
for substantial future dialogue with private and public sector investors.
The three final evaluation reports will also have annexes that include current status
project inputs and outputs if these are not readily indicated in the body of the report.
Other required annexes to the reports are: technical and management issues raised
during assessment requiring elaboration, the project evaluation scope of work, a
description of the methodology used in assessment, bibliography of documents
reviewed and a list of agencies contacted, individuals interviewed and other relevant
information.
In addition to the three final project evaluation reports, ten (10) copies of a stand￾alone report synthesizing CLUSA, CARE, and RNRMP/ADMADE lessons learned
that have applicability to food security, rural income and community natural resource
conservation is also required. This report will include an appropriate introduction
describing the document’s contents, a main body laying out lessons learned from the
three project interventions that have relevance to ongoing or future food security,
rural income or natural resource conservation activities, and a concluding chapter
containing recommendations on how lessons learned can be disseminated to
beneficially influence future investments. Again, the attractiveness and user friendly￾ness of this report is key.
Article 7. Relationships and Responsibilities
The Contractor shall perform the tasks described above under the general guidance of
David Soroko, SO1 Team Leader. The consultancy team will work closely with
USAID activity managers involved with the individual projects.