PSNP PLUS

LINKING POOR RURAL HOUSEHOLDS TO MICROFINANCE AND MARKETS

Impact Assessment of Livestock Value Chain Interventions Final Impact Assessment of the PSNP Plus in Raya Azebo

November 2011

John Burns and Solomon Bogale

! "#$ "#%&'(()&)*)+!&,(*-.,(!/-& &0(&1"2)(3"#+#2.&+#/&1+)$.0- & &

!"#$%&'()!"($**+(%&(+,+-#(")!%*$%-('+-.,-)"(/+*0($+*.,1)%2"(3-/")4(&")($+*'+-0(

566785699(

'-#'&!"#$&

2

3

Acknowledgements The PSNP Plus Program “Linking Poor Rural Households to Microfinance and Markets in Ethiopia’ was funded by the United States Agency for International Development. The Program was led by CARE, and implemented by CARE, Save the Children (UK), Catholic Relief Services (CRS) and partners, and the Relief Society of Tigray (REST). The Netherlands Development Organizations (SNV) and the Feinstein International Center, Tufts University provided technical support to the implementing partners. We would like to acknowledge the contributions of the field research team Tikabo Gebre Mariam and Girmay Murutse from Mekele University, and Tesfay Hagos and Abraham Haftu from Tigray Agricultural Research Institute. Without their efforts this study would not have been possible. We are also grateful to the Raya Azebo woreda administration for supporting the study. We would also like to thank the Relief Society of Tigray (REST) offices in Mekele and Mehoni for supporting the assessment. In particular, our gratitude goes to Dr. Mulugeta Berhanu, Dawit Libanos, Zeratsion Fesseha, Yishak Desta, Teklehaymanot Etsana and Birikti G/Egziabher. We would also like to thank the woreda officials, development agents and respondent farmers in Raya Azebo for their support and participation. This report and the associated study were made possible by the generous support of the American people through the United States Agency for International Development (USAID). The contents are the responsibility of CARE and its PSNP PLUS partners and do not necessarily reflect the views of USAID or the United States Government. The content of this report is derived from research carried out by the Feinstein International Center, Tufts University under the USAID funded PSNP Plus project. The contents of this report have not been endorsed by the other PSNP Plus partners, and do not necessarily reflect the views of these organizations.

4

Contents

SUMMARY .............................................................................................................................................. 7  1.   INTRODUCTION ........................................................................................................................... 10  

1.1 Background to the PSNP Plus Program ..................................................................................... 10  1.2 PSNP Plus Overview .................................................................................................................. 11  1.3 Background to the Study in Raya ................................................................................................ 14  

1.3.1 Study Area General Characteristics .................................................................................... 14  1.3.2 PSNP Plus Interventions in Raya ........................................................................................ 14  

2.   ASSESMENT METHODOLOGY ................................................................................................... 17  2.1   Study Design ........................................................................................................................... 17  

2.1.1   Research Questions ........................................................................................................ 17  2.1.2 Study Components .............................................................................................................. 17  2.1.3 Indicator Selection ............................................................................................................... 18  

2.2   Sampling ................................................................................................................................. 18  2.2.1   Method and Size .............................................................................................................. 18  2.2.2   Study Locations ............................................................................................................... 19  

2.3   Data Collection Methods ......................................................................................................... 20  2.3.1   Household Interviews ...................................................................................................... 20  2.3.2 Focus Group Methods ......................................................................................................... 21  

2.4   Data Analysis .......................................................................................................................... 21  3.   RESULTS ...................................................................................................................................... 22  

3.1   Context and Background ......................................................................................................... 22  3.2 Income ........................................................................................................................................ 23  

3.2.1 Income Sources ................................................................................................................... 23  3.3 Expenditure ............................................................................................................................. 25  

3.4 Credit and Savings ...................................................................................................................... 27  3.5 Asset Levels and Changes ......................................................................................................... 28  

3.5.1 Land Holdings ...................................................................................................................... 28  3.5.2 Livestock holdings ............................................................................................................... 29  3.5.3 Productive Assets/Tools ...................................................................................................... 31  3.5.4 Household Items .................................................................................................................. 32  

3.6 Changes in Household Food Security ......................................................................................... 33  3.7 Value Chain Production and Sales ............................................................................................. 34  3.8 Scoring of Project Benefits .......................................................................................................... 36  3.9 Strengths and Weaknesses of Program Interventions ................................................................ 37  

4. DISCUSSION .................................................................................................................................... 39  4.1 Methodological Issues ................................................................................................................. 39  4.2 Income and Expenditure ............................................................................................................. 40  

4.2.1 Income Sources ................................................................................................................... 40  4.2.2 Actual Income and Relative Expenditure ............................................................................. 41  

4.3 Credit and Savings ...................................................................................................................... 41  4.4 Changes in Assets ...................................................................................................................... 42  

4.4.1 Land Holdings ...................................................................................................................... 42  4.4.2 Livestock Holdings ............................................................................................................... 42  4.4.3 Productive Assets and Household Items ............................................................................. 43  

4.5 Changes in Food Security ........................................................................................................... 43  4.6 Utilization of Project Derived Income and Credit ......................................................................... 43  

4.6.1 Savings and Loan Utilization ............................................................................................... 43  4.6.2 Utilization of Income from Livestock Fattening .................................................................... 44  

CONCLUSION ....................................................................................................................................... 44  ANNEXES ............................................................................................................................................. 49  

Annex I Community wealth indicators (n=27 Groups) ....................................................................... 49  Annex II Ranking of Reasons for an Increase in Land Holdings ....................................................... 50  Annex III Household Checklist .......................................................................................................... 51  

5

List of Tables Table 1: Value chain outputs under PSNP Plus ..................................................................... 13  Table 2 Sampling frame and actual sample ........................................................................... 18  Table 3: Final assessment coverage ...................................................................................... 19  Table 4 Summary of household questionnaire themes and methods .................................... 20  Table 5: Project Intervention timeline ..................................................................................... 22  Table 6: Characteristics and background data on assessment participants .......................... 23  Table 7: Perceived changes in household income since 2009 .............................................. 25  Table 8: Factors contributing to an increase in income .......................................................... 25  Table 9 Changes in the value of savings and loans ............................................................... 27  Table 10: Changes in land holdings ....................................................................................... 28  Table 11 Food Security Calendar (n= 15 groups) .................................................................. 33  Table 12 Factors contributing to improvements in food security ............................................ 33  Table 13: Livestock value chain transfers and sales .............................................................. 34  Table 14: Scoring of other project outcomes (Sheep) ............................................................ 36  Table 15: Scoring of other project outcomes (Cattle) ............................................................. 36  Table 16 Livestock Value Chain SWOT Analysis (n=27 groups) ........................................... 37  Table 17 Village Saving and Loan Association SWOT analysis (n=5 groups) ....................... 38   List of Figures Figure 1: Changes in relative contributions of different income sources (Sheep) .................. 23  Figure 2: Changes in relative contributions of different income sources (Cattle) ................... 24  Figure 3: Relative Expenditure on Key Items (Sheep) ........................................................... 25  Figure 4: Relative Expenditure on Key Items (Cattle) ............................................................ 26  Figure 5: Utilization of savings and credit ............................................................................... 28  Figure 6: Changes in livestock holdings (Sheep) ................................................................... 29  Figure 7: Changes in livestock holdings (Cattle) .................................................................... 29  Figure 8: Changes in productive assets (Sheep) ................................................................... 31  Figure 9: Changes in productive assets (Cattle) .................................................................... 31  Figure 10: Changes in household items (Sheep) ................................................................... 32  Figure 11: Changes in household items (Cattle) .................................................................... 32  Figure 12: Utilization of income from sheep fattening ............................................................ 34  Figure 13 Utilization of income from cattle fattening .............................................................. 35  

6

Acronyms and Abbreviations AARC Alamata Agricultural Research Center ACSI Amhara Credit and Saving Institute Agric Inputs Agriculture/farming Inputs CARE Cooperative for Assistance and Relief Everywhere CRS Catholic Relief Services DECSI Dedebit Credit and Saving Institution DPPA Disaster Preparedness and Prevention Agency ETB Ethiopian Birr FGD Focus Group Discussion FSP Food Security Program GFDRE Government of the Federal Democratic Republic of Ethiopia GRAD Graduation with Resilience to Achieve Sustainable Development HABP Household Asset Building Program HH Household HI Home Improvement IGA Income Generating Activities LIS Longitudinal Impact Study LIU Livelihoods Information Unit (DPPA) M&E Monitoring and Evaluation MDTCS Micro Development Training and Consultancy Services MFI Micro Finance Institute MoARD Ministry of Agriculture and Rural Development OCSSCO Oromia Credit and Saving Share Company OFSP Other Food Security Programs PSNP Productive Safety Net Program PSNP-PIM PSNP Program Implementation Manual PSNP Plus Linking Poor Rural Households to Microfinance & Markets (Project) P.Trade Petty Trade Qty Quantity Radio/CP Radio/Cassette Player REST Relief Society of Tigray RFA Request for Applications SCUK Save the Children Fund (UK) SILC Saving and Internal Lending Committee SNV Netherlands Development Organization Social Oblig. Social Obligations SR Small Ruminant SPSS Statistics Package for Social Sciences SWOT (analysis) Strengths Weaknesses’ Opportunities Threats TARI Tigray Agricultural Research Institute Trad Traditional (beehives) USAID United States Agency for International Development VC Value Chain VSLA Village Savings and Loan Association

7

SUMMARY

The USAID funded PSNP Plus program ‘Linking Poor Rural Households to Microfinance and Markets in Ethiopia’ was launched in late 2008 and will run until December 2011. The PSNP Plus was designed as a three-year program in support of the Government of Ethiopia’s Productive safety Net Program (PSNP) which provides food and or cash to chronically food insecure households in exchange for labor on rural infrastructure projects, or direct transfers to households unable to participate in physical labor activities. A consortium of six international and national NGO’s is implementing the program led by CARE. The program was initially implemented in nine pilot woredas in Tigray, Amhara, Oromia and Dire Dawa regional states, with the overall goal of building household resilience and household assets through market linkages and access to microfinance. This goal is directly linked to the objective of facilitating the graduation of households from the PSNP and out of chronic food insecurity. Since it was launched, the program has been linking PSNP households to both formal and informal microfinance. These interventions have included the establishment of Village Saving and Loan Associations (VSLA), and the provision of credit for agricultural inputs. Complementary to these activities, the program has been linking participating households to market opportunities by supporting the development of livestock, cereal, white pea bean and honey value chains. Ultimately, the combination of the programs microfinance and value chain interventions is expected to contribute towards livelihoods diversification, household resilience, and an increase in household income and assets with associated improvements in PSNP graduation. These outcomes and impacts are reflected in the programs causal model, which in summary proposes that increased access to markets and the enhanced use of microfinance leads to asset accumulation and improvements in PSNP graduation. In order to test this causal logic, a longitudinal impact study (LIS) was included under one of the programs strategic objectives. The LIS included a baseline mid-term and final impact assessment in four of the program study areas. This report presents the results from the final impact assessment of the program in Raya Azebo woreda, implemented by the Relief Society of Tigray. The study used a pre-post test design with controls, and focused on assessing the livestock credit and cattle and small ruminant value chains being implemented under the project. The final assessment was carried out from August to September 2011 during the final year of project implementation. The assessment results indicate that the project has had a positive impact on the livelihoods of participating households. Although graduation levels remain low, the project has contributed to an increase in income and assets and improvements in household food security. The project also appears to have been instrumental in helping households protect their livestock assets. The study findings show that there has been a significant increase in the contribution of income derived from livestock fattening relative to all other income sources for both cattle and small ruminant (sheep) value chain participants. The results also show that assessment participants have experienced an overall increase in income since the project started with project participants experiencing a significantly greater percentage increase in income than control group participants. Respondents attributed this improvement to a number of reasons with the livestock value chains being identified and scored as an important contributing factor. At the time of the assessment, the mean income from the sale of fattened sheep was estimated at 1,425 Ethiopian birr and the mean income from the sale of fattened cattle was estimated at 2,476 birr for participating households. This project-derived income was utilized

8

on a number of livelihoods activities with livestock investments being the most important expenditure for participants in both value chains. For example, sheep value chain participants’ re-invested 346 birr (mean expenditure) in livestock, and cattle value chain participants’ re-invested 1,432 birr. Other expenditures included food purchases and the repayment of loans and debts. There has also been a significant increase in the value of savings for project participants relative to both the baseline and the control. These savings not only represent an important financial asset in terms of future investments, but also provide insurance against livelihoods shocks and might be considered a useful proxy for resiliency. The results also indicate that credit and savings have contributed towards improved food security and asset accumulation for project participants. For example, mean household savings and loans combined for the past year was estimated at 2,535 Ethiopian birr for the sheep sample and 1,763 birr for the cattle sample. This money was invested in a variety of livelihoods activities with the most important expenditures including investments in livestock, land and food purchases. The project has also had a positive impact on other types of assets including land, livestock, tools and household items. The results show that there has been a significant increase in land holdings for project participants since the project started. Land holdings are typically measured in terms of utilization, which is associated with household labor capacity and ownership of draft animals, as well as the ability to employ farm labor. These results therefore suggest that household productive capacity has improved greatly since 2008. The results show no change in land holdings for control group participants. The results show no significant change in livestock assets for the cattle value chain participants and an increase in calves and a decrease in cows for the sheep sample. However, the results from the midterm assessment showed a significant decrease in certain livestock assets due to stress sales and mortality associated with the drought in 2009. The final results indicate that a considerable recovery of livestock assets has taken place for project participants since the midterm assessment, whereas this has not been the case for control group participants. This would suggest that the project has contributed to post drought asset recovery and that project participants have been better able to protect and maintain their livestock in comparison to non-project participants. Consistent with this, the projects role in helping people protect their assets was scored as the second most important project benefit by assessment participants. Since the project started both the intervention and control group participants have experienced a significant increase in certain types of productive assets and household durables. For certain items such as pickaxes, mats, cups and Jeri cans, this increase has been significantly greater for project participants than for the control group. There have also been considerable improvements in household food security since the project started. Although in part this can be attributed to non-project factors such as improved rainfall and the PSNP, the projects contribution has also been important. For example, the projects role in contributing to household food security was scored as the single most important project benefit by value chain participants. Similarly, focus group participants ranked livestock credit as one of the most important factors contributing to improvements in food security. At the time of the assessment, very few assessment participants had graduated from the safety net program, possibly because PSNP graduation benchmarks had not been met or indeed assessed. Nonetheless, the results from this case study indicate that microfinance and value chain interventions can have a significant impact on the livelihoods of PSNP participants. The

9

assessed changes in financial assets and improvements in food security certainly suggest that well designed and implemented livestock value chains in concert with the provision of credit may represent a potential pathway to PSNP graduation in similar highland contexts.

10

1. INTRODUCTION

1.1  Background  to  the  PSNP  Plus  Program  

In recent years considerable progress has been made in addressing chronic food insecurity and the risk of repeated crisis in Ethiopia, much of this can be attributed to the Government of Ethiopia’s Food Security Program (FSP). The overall goal of the FSP is to attain food security for both chronically and transitory food insecure households in rural Ethiopia (MoARD, 2009). When it was launched in 2005, the program was built around three key components, a Productive Safety Net Program (PSNP), a Voluntary Resettlement Program and Other Food Security Programs (OFSP). Under the 2010-2015 FSP program, a Capital Intensive Community Infrastructure Component has been included, and the OFSP has been replaced with the Household Asset Building Program (HABP). The PSNP component of the FSP was specifically designed to assist chronically or ‘predictably’ food insecure households as opposed to households affected by transitory food deficits. The program provides either cash or food in exchange for labor on rural infrastructure projects, or direct cash and food transfers for households unable to participate in physical labor. The primary objectives of the PSNP are therefore to prevent chronically food insecure households from selling their assets during times of drought, and to build community assets through involving these households in public works programs (Pankhurst, 2009). In 2006 the PSNP provided support to an estimated 8.6 million people, making it the second largest social transfer program in Africa (Sabates-Wheeler and Devereux, 2010). Ultimately participating households are expected to ‘graduate’ from the PSNP and out of chronic food insecurity. However, although the PSNP has had a significant impact on smoothing consumption, and protecting the assets of the chronically food insecure (Sharp et al. 2006, Devereux et al, 2006), little progress has been made in terms of graduating households from the program (MoARD, 2009). Various definitions for graduation have been proposed, most of these involve the concept of households moving out of chronic food insecurity (for example see, PSNP-PIM, 2006, Slater et al, 2006, and Devereux et al, 2006). A PSNP graduation guidance note defines graduation as follows (MoARD, 2007: 2): “A household has graduated when, in the absence of receiving PSNP transfers, it can meet its food needs for all 12 months and is able to withstand modest shocks. This state is described as being food sufficient”. Essentially graduation involves a two-stage process: the first stage is graduation from the PSNP program, and the second stage involves graduation from the Food Security Program. In order for households to graduate it is recognized that they need to be linked to Other Food Security Programs that go beyond the PSNP food and cash safety net transfers (MoARD, 2006). The OFSP include interventions that provide credit and loans for agriculture as well as non-farm income generating activities, and the provision of ‘agricultural technologies’ such as extension services, and inputs (Gilligan et al, 2009). While the overall goal of the PSNP is to address food insecurity through household asset protection and community asset creation, the OFSP were designed to increase participant’s income from agricultural production, and build up household assets (Gilligan et al, 2009). In theory, this accumulation of income and assets enables households to graduate from food insecurity and out of the PSNP. In recognition of this, and in support of the Government of Ethiopia’s FSP, in 2008, USAID issued a $ US 12,000,000 Request for Proposals (RFA) entitled “Linking Poor Rural Households to Microfinance and Markets in Ethiopia”. The RFA

11

was launched with the objective of demonstrating that the “adoption of market –led livelihood options for the persistently poor through sustainable links to markets and microfinance services” results “in increased assets at the household level and therefore more resilient households” (USAID, 2008: 18). The RFA also suggests that the value chain approach be considered as an appropriate methodology for linking poor households to markets. More specifically, the RFA called for projects that would contribute to the following higher goals (USAID, 2008: 18-19): • Reduced food insecurity and improved resiliency in vulnerable households • Increased rural economic growth opportunities for the poor to diversify livelihoods • Demonstrate a new market-driven approach to poverty reduction in Ethiopia • Expanded adoption and scaling up of market-driven approaches by new actors such as

the Government of the Federal Democratic Republic of Ethiopia (GFDRE) • Improved access to microfinance services through a graduated assistance program

The RFA also required that proposals demonstrate how project results, outcomes, and the ‘replicability’ and sustainability of interventions would be measured and documented. Consistent with this, the RFA called for a preliminary causal model presenting the logic of how the project would achieve the desired outputs, outcomes and impacts, and how these would be measured (USAID, 2008). In response to the RFA, the PSNP Plus program was specifically designed to provide alternative and accelerated pathways to PSNP graduation for chronically food insecure households lacking access to other food security and microfinance interventions (PSNP Plus, 2008).

1.2 PSNP Plus Overview

The original PSNP Plus program was launched in late 2008 as a three-year program implemented by a consortium of six international and national NGO’s led by CARE. The program was initially implemented in nine pilot woredas in Tigray, Amhara, Oromia and Dire Dawa regional states. The program strategy is based on a push-pull causal model designed to link 42,414 participating households to microfinance and markets through the provision and integration of contextually relevant financial services and value chain interventions. The proposed causal model is based on the assumption that ‘improved linkages between poor households and commodity markets, plus enhanced use of microfinance services leads to asset accumulation at household level with associated improvements in PSNP graduation and resilience’ (CARE, 2008). Consistent with this causal logic, the overall program goal states that: “Targeted PSNP households’ resiliency improved and livelihood assets enhanced as a means towards achieving graduation.” In order to achieve this goal, the program was structured around three interlinked objectives designed to bring immediate positive impact to participants (PSNP Plus, 2008). Objective 1: Targeted PSNP households have increased their financial assets as a result of access to financial products and services. Objective 2: Targeted PSNP households are engaged in functioning markets. Objective 3: Government and private sector strategies show greater support for engaging PSNP participants in market-based activities.

12

Under the first objective, the program provides both formal and informal financial products and services such as credit and savings. Under the second objective, the program has been supporting value chain interventions in honey, white pea beans, cereals and livestock. The combination of these interventions was specifically designed to enable participants to enter markets and accumulate assets, with the expectation that after three years eighty percent of PSNP Plus households would meet the criteria for PSNP graduation (PSNP Plus, 2008).

The provision of financial services is ultimately geared towards assisting people in building up assets by utilizing loans and savings to invest in high return productive and other income generating activities. For example, the program is designed to assist participants in securing loans for value chain inputs. In the absence of formal microfinance, the program has been promoting the establishment of Village Saving and Loan Associations (VSLA) alternatively called Saving and Internal Lending Committees’ (SILC) based on the village agent model developed by CARE. This approach typically involves a group of between 10-25 members. The project provides training and resources to these groups to enable them to manage, maintain and increase their own financial assets such as savings and loans. Under the VSLA approach, members use their own cash resources to lend funds to one another, charge an acceptable interest rate, and re-lend funds on a rotating basis. The program strategy also involves using these groups as a vehicle to link members to formal microfinance. By demonstrating that group members’ financial literacy and knowledge on savings increases over time, the project aims to convince MFI’s to accept groups and individuals as clients. As such, the SILC groups are intended as a catalyst to provide the linkage between informal and formal microfinance (MDTCS, 2010). Under the market linkage component, the program has been supporting four-commodity value chains viz. livestock, honey, white pea beans, and cereals. Among other criteria, the value chains were selected by consortium partners based on the anticipated production potential of these commodities in the project area, income earning potential, and market potential in terms of demand and growth. The program aims to assist PSNP Plus participants in the production and marketing of these commodities. On the supply side, the objective of these interventions is not only to increase production, but also to improve the quality of these products with a view to adding to their market value. On the production side, the program provides technical support such as training, as well as certain types of specific inputs such as honey production accessories, livestock and improved seed varieties. The training components and transfer of inputs is facilitated through producer or marketing associations established under the program. The production side of the value chains is also complemented by the microfinance component, in that production inputs such as seeds, livestock and beehives are supplied to project participants on a credit basis from formal microfinance institutions such as Oromia Credit and Saving Share Company (OCSSCO), Dedebit Credit and Saving Institution (DECSI) and Amhara Credit and Saving Institution (ACSI). Under the market linkage component the program has been establishing facilities such as storage and collection centers to prevent spoilage and facilitate marketing, linking farmers to government extension services and the private sector, and establishing market information platforms. Table (1) gives a summary of the objectives and expected outputs of the value chain activities.

13

Table 1: Value chain outputs under PSNP Plus

Objectives Expected Outputs Critical bottlenecks for each value chain inhibiting PSNP household’s entry to value chain identified.

• Existing value chain assessments updated and new value chains validated.

Targeted PSNP households start production or improve productivity and quality of selected products.

• Targeted PSNP households have formed producer or marketing associations.

• Newly formed producer or marketing associations have access to production inputs.

• Targeted PSNP households received training or technical assistance on productivity and quality of production.

• Government, private sector, research institutions and others are providing targeted PSNP households with market extension services, post-harvest storage, assistance with handling and marketing.

• Women have the skills necessary to be successful entrepreneurs.

• Private sector engaged in value chain activities and linkages based on market demand created.

• Private sector and producer/marketing associations engaged in contracts, trader credit, warehouse receipt schemes and other contract farming.

Stakeholder forums and coordination groups help value chain actors and stakeholders resolve problems and meet shared goals.

• Coordination group and stakeholder forums established for value chain development.

Market information platforms provide targeted producers with the information necessary to negotiate fair prices, access to technical assistance and productive inputs.

• Market information platforms created.

Source: PSNP Plus Project Proposal (2008)

In line with the PSNP Plus being a pilot program, a specific learning component was incorporated under objective number 3, with a view to generating evidence to influence key stakeholders on how combinations of microfinance and market oriented interventions can enhance PSNP graduation, an independent Longitudinal Impact Study (LIS) was included as

14

a specific program activity. The study ultimately seeks to test and validate the programs causal model by assessing whether the strategies and activities implemented under Objectives 1 and 2 do indeed result in asset accumulation and more resilient households. The LIS was carried out in four study areas covering each of the regions represented by the program, and all four of the PSNP Plus value chains. The study tracked and assessed changes in household assets over three points in time (baseline, mid-term, end of program) using a variety of research designs and methods. This report presents the findings from the final impact assessment of the PSNP Plus project in Raya Azebo (Raya), which took place from August 10th to September 6th 2011.

1.3 Background to the Study in Raya

1.3.1  Study  Area  General  Characteristics   Raya Azebo is one of 36 woreda’s in Tigray, with an estimated population of 105,298 people (DPPA, 2007). Raya Azebo shares a border with Afar region to the east, and Amhara region to the south. The woreda is situated in the ‘Raya Valley Sorghum and Teff’ Livelihood zone, characterized by undulating mountains, plains and rugged terrain (DPPA, 2007). An estimated 88% of the woreda’s population lives in rural areas (SNV, 2010). As such, agricultural production is the main livelihoods activity, and involves a mixture of food and cash crop production (DPPA, 2007). Although annual rainfall is moderate, ranging from 450-600 mm, the availability of sufficient farmland and fertile clay/loam soils makes the area well suited to crop production (DPPA, 2007:4). The main crops grown are maize, sorghum and teff, and draft animals are extensively utilized for land preparation (DPPA, 2007). Agricultural production is dependent on the Belg and Kiremti rains, and although a crop surplus can be expected when there is good rain, production is often affected by unreliable or erratic rainfall, as well as a variety of crop pests (DPPA, 2007). Livestock production and trading are also important livelihoods and economic activities in the area. This includes rearing of both cattle and small ruminants. There is also a healthy livestock trade in the area, and markets in the woreda such as Kukufto, receive a considerable amount of livestock traffic, including camels in transit from Afar to Sudan. Other economic activities include petty trading and agriculture labor (DPPA, 2007).

1.3.2 PSNP Plus Interventions in Raya

Under the microfinance component, the project aims to improve participant’s access to financial products and services, by linking participants to formal microfinance institutions (MFI). The project in Raya also includes an informal microfinance component involving a small pilot to test the Village Saving and Loan Association (VSLA) approach in the area. In Raya, the project provides credit through MFI’s for value chain commodities such as sheep and cattle and other inputs. It is anticipated that over time, PSNP Plus participants will also be linked to other financial services such as credit and loan facilities. The main MFI in Raya Azebo is Dedebit Credit and Saving Institution (DECSI), with two offices in the woreda. DECSI was established in response to the work done by REST in providing credit services to the poor, and since it was officially registered in 1997, much of DECSI’s work has continued to focus on the provision of credit services for the poor in the rural areas of Tigray. Although the PSNP Plus project recognizes the challenges of providing financial services to the poor, DECSI’s experience and success provides a promising example of what can be achieved in the Ethiopian context (see Borchgrevink et al, 2003 and

15

Borchgrevink et al, 2005). The project has also been piloting an informal credit and saving component for 300 participants based on the VSLA approach. Under the market linkage component, REST is supporting three commodity value chains in Raya Azebo. This includes two livestock fattening value chains, cattle fattening and small ruminant fattening, and a cereal (teff) value chain. According to the original proposal, the project aims to support 2,014 households under the two livestock fattening chains, and 3,000 households under the cereal value chain (PSNP Plus, 2008). Under the two livestock value chain interventions, the project aims to increase the income derived from livestock sales for project participants, and to diversify people’s income and reduce their dependency on crop production and sales (SNV, 2010). The implementation of activities for the fattening value chains started in the middle of 2009. During this period, participants received a five days training on different topics, including animal selection and fattening techniques, animal management (feeding, housing and disease control), livestock marketing and business management, financial literacy and record keeping. The transfer of assets was carried out by a committee composed of representatives from REST, the woreda office of agriculture and rural development, the office of finance and economic development, development agents and community representatives. Cattle and sheep were then purchased and distributed to selected participants on a credit basis, with loans being provided to purchase one ox (cattle value chain) or three to six small ruminants (sheep value chain), with the expectation that the loans be repaid with interest once the livestock have been fattened and sold (SNV, 2010). At the beginning of the project cattle value chain participants received a loan of 2000 birr, and sheep value chain participants received 1000 birr. Participants from the cattle fattening value chain were also loaned 400 birr to be used for supplementary feed. Over time this credit increased to accommodate increasing livestock prices and inflation, with cattle participants receiving as much as 31,00 birr and sheep participants receiving 1,300 birr. The project design also includes a revolving fund mechanism, whereby once the loans have been repaid; the funds are then lent to other households (SNV, 2010). During the first year of the project, 124 cattle and 1,242 small ruminants were transferred (CARE, 2010 a). By March 2010 over 50% of the cattle had been sold for between 2500-300 birr, and over 62% of the small ruminants for between 270-450 birr (CARE, 2010 a). Project documents indicate that all of the remaining animals would have been sold by the early 2010 (CARE, 2010 a). The selection of project areas in the woreda was based on the availability of irrigation and forage, and above average market access (SNV, 2010). Amongst other factors, project participants were selected based on vulnerability criteria, willingness to participate, prior experience with fattening, and the amount of land they had to produce forage (SNV, 2010). Under the informal microfinance component, seven VSLA groups were established in Raya during the first year of the project. Six of these were located in Mechare tabia, and one in Hujera village of Were Baye tabia. At the time of the midterm assessment, group membership ranged between 13-27 participants in the six groups assessed. Group members received training on the VSLA approach, financial literacy, record keeping, and income generating activities.

16

Since it was launched, the project in Raya has faced a number of challenges. The drought in 2008-2009 had a negative impact on income and asset accumulation for project participants. For the livestock value chain participants, the issue of feed shortage and poor animal health associated with drought limited the impact of the project during the first year of implementation. In response, a study was carried out to identify new drought resistant feed and fodder technologies (CARE, 2010). In addition to this, the introduction of credit in Raya Azebo has been particularly challenging. Roughly 40% of the population in the woreda is Muslim, and local interpretation of Islamic law prohibits Muslims from engaging in money lending or interest earning activities (Getahun, 2010). Clearly this has had a number of implications on both the microfinance and value chain components that define the project. Seeing as one of the key obstacles appeared to be the use of the word ‘interest’, REST provided value chain assets in kind, with a view to charging recipients with a service charge once these animals were sold (Getahun, 2010). However, religious leaders did not accept the difference between ‘service charge’ and ‘interest’, and compelled project participants to either return the assets they received, or be subjected to various types of social exclusion (Getahun, 2010). In response to this challenge, REST has worked with the community, Muslim leaders and Islamic scholars to address this issue (Getahun, 2010). Drawing upon examples of Islamic banking systems elsewhere, a new approach (Murabaha) was agreed upon, whereby the interest or fee are considered as a service charge or commission, and lumped together with the loan amount from the outset (Getahun, 2010). At the time of the mid term assessment, this new approach appeared to be working quite well although there was still some reluctance on the part of some community members. These challenges aside, project implementation from planning to the delivery of inputs, has been exceptionally fast in comparison to other PSNP Plus areas. For example, in the first year of implementation, the project achieved 100% of its annual performance targets for the cattle value chain and 69% for the small ruminant value chain (REST, 2009).

17

2. ASSESMENT METHODOLOGY

2.1 Study Design

2.1.1 Research Questions The LIS was designed to test the programs causal model and validate whether improved linkages between poor households and commodity markets, plus enhanced use of microfinance leads to asset accumulation at the household level with associated improvements in PSNP graduation. In order to test this model, the final assessment tracked changes in household assets across three points in time (baseline, mid-term and final) using the same household participants for each stage of the study. The final assessment essentially set out to answer the following key research questions: 1. What changes in household assets has occurred since the project started? 2. What factors contributed to any assessed change in these assets? 3. What was the relative contribution of project factors to any assessed change? The study focused on measuring changes in physical and financial assets such as land, livestock and productive assets, these being benchmarks for PSNP graduation. The study in Raya focused on the small ruminant and cattle fattening value chain activities implemented under the PSNP Plus. The study also assessed changes in food security and income, and the relative contribution of different income sources. This was done with a view to capturing the relative impact of the project value chains on household income, and to capture livelihoods diversification, which might be considered a useful proxy for resilience. Actual changes in certain key investments and expenditures were also measured as a proxy for real income, and to capture investments in livelihoods assets such as health, education, livestock and farming inputs. More specifically, the study measured actual project-derived income from both credit and savings, and value chain profits, and assessed the utilization of this income as an alternative way of measuring impact.

2.1.2 Study Components The assessment in Raya used a pre-post test design with controls to assess livelihood changes of households participating in the programs livestock value chain interventions. For the purpose of this study, impact is broadly defined in terms of significant and measurable changes that have taken place since the program started that can be attributed to the programs microfinance and value chain interventions. The study also considers impact in terms of the utilization of project transfers such as credit and project derived income and the livelihoods investments and benefits obtained in relation to these. There were two main components to the final impact assessment in Raya, household interviews and focus group discussions. As implied, the household component used an individual household as the unit of analysis - the household also being the unit for PSNP participation and graduation. The household component was designed to collect mostly

18

quantitative data using a conventional questionnaire format, and including a number of standardized participatory assessment methods. The focus group component was designed to collect mostly qualitative contextual data on the project activities, communities, and areas. The focus group discussions were structured around a set of standardized participatory assessment tools providing some complementary numerical data. A number of key informant interviews were also carried out. These were used to collect secondary data on the project and study area.

2.1.3 Indicator Selection The choice of indicators was largely based on PSNP graduation benchmarks at the time of the baseline assessment. A number of additional indicators were also collected during pre-baseline scoping visits to the study area. These included additional indicators on assets, sources of income, and common household expenditures. These indicators were then further refined during the pre-testing for the baseline assessment, and validated during the actual baseline study.

2.2 Sampling

2.2.1 Method and Size For the household component of the study, simple random sampling was used for the small ruminant value chain sample. Every registered participant was considered, and the latest project registration list was used as the sampling frame. Out of this sampling frame, a total of 150 households were randomly selected for the baseline assessment. The same approach was used for the cattle value chain. However, seeing as only seven female participants were involved in the cattle value chain, these seven women were purposively selected, and 80 male participants were randomly selected for the baseline assessment. The PSNP participant lists provided the sampling frame for the comparison (control) group sample, but excluding households involved in PSNP plus project activities. However, respondents were purposively selected based on their willingness and availability to participate in the study. The actual sample for the baseline assessment was then used as the sampling frame for the final assessment. Table 2 provides a summary of the planned and actual samples for the baseline and final assessments. Table 2 Sampling frame and actual sample

Sheep Value Chain Control Cattle Value Chain Sampling Frame 339 NA 124 Planned Sample 150 NA 87 Actual Sample Baseline 136 130 82 Actual Sample Final 116 105 77 A total of 42 focus group discussions were carried out across the study area. These exercises included participants from both the projects livestock value chains and non-project PSNP participants. Participants from six of the existing VSLA groups also participated in some of these focus groups. Participation in these discussions was voluntary, however

19

attendance by project participants had been requested in advance. There was also no overlap between participants selected for household interviews, and those involved in the focus groups. A number of key informant interviews were also carried out with project staff, woreda officials, and development agents. These interviews were used to collect background information, triangulate findings from household and focus groups exercises, and to fill in gaps.

2.2.2 Study Locations The assessment team visited all 8 tabia’s in Raya Azebo where the livestock value chains had been implemented at the time of the baseline assessment. The team also visited an additional tabia (Mechare) where a VSLA group had been established under the project. Table 3 provides a summary of the geographical coverage for the final assessment. Table 3: Final assessment coverage

No. Kebele Category (# of Households) Assessment Method

Cattle VC Sheep VC Total Household Interview

Focus Groups

1 Bageai 11 0 11 P P 2 Bala 0 24 25 P - 3 Erba 0 5 5 P P 4 Genetie 13 26 37 P P 5 Horda 0 0 0 - P 6 Kara adsisho 14 29 43 P P 7 Kukufto 18 0 18 P P 8 Maru 0 5 5 P P 9 Mechare 0 0 0 - P

10 Wargiba 0 0 0 - P 11 Were Baye 21 27 48 P P

Total 76 116 192 P P

20

2.3 Data Collection Methods

2.3.1 Household Interviews The interviews for the household component were carried out by a team of four data collectors under the supervision of an assessment coordinator. These interviews were carried out on an individual basis using a standardized questionnaire that included a number of participatory exercises and some qualitative data. The assessment tools were field tested with non-assessment participants and refined shortly before the actual assessment began. The household questionnaire for the final assessment included the following themes: Table 4 Summary of household questionnaire themes and methods

Section/Theme Types of Information Collected (method) Sample Size 1 Household and

Project Background Information

• PSNP and PSNP Plus activities and participation • Household (HH) education levels • Recent livelihoods shocks experienced • HH PSNP graduation status • Factors contributing to PSNP graduation (simple ranking)

193

15

2 Savings and Loan Information

• Recent HH savings history • Recent HH borrowing history and source of loans • Utilization of HH savings and loans

298

3 Asset Inventory

• Current land holdings • Current livestock holdings • Livestock sales and mortality • Current levels of productive assets (tools) and HH items

298

4 Income • Relative contribution of different income sources (proportional piling using 100 counters)

• Perceived changes in actual income (proportional piling against a nominal baseline of 10 counters)

• Scoring of factors contributing to an increase in income (proportional scoring using 50 counters)

298

246

5 Livestock sales • Livestock fattening and sales 2011 • Income utilization from fattening 2011

193

6 Expenditure • Relative expenditure on key items (proportional piling using 100 counters)

298

7 Other benefits (outcomes)

• Scoring of PSNP Plus programs anticipated outcomes (simple ranking and scoring)

193

See Annex III for the complete household questionnaire

21

2.3.2 Focus Group Methods The focus group component of the final assessment used a mixture of qualitative, quantitative, and participatory data collection methods. These discussions were primarily used to collect descriptive contextual information on the PSNP Plus program and more general information on the project area. The focus groups were structured around a checklist, which included a set of standardized participatory exercises. These included a SWOT (Strengths, Weaknesses, Opportunities, Threats) analysis of the different program interventions being assessed, and a food security scoring exercise to assess temporal changes in food security.

2.4 Data Analysis The household results were analyzed collectively with the baseline results, and summarized using Statistical Packages for Social Science (SPSS version 19). Comparisons between the baseline and final, and the treatment and control samples were done to assess changes in; income, income sources, assets, expenditure, and savings and loan values by source with mean values being calculated at ninety five percent confidence interval using SPSS. Livestock production and sales and factors contributing towards an increase in income were also calculated at ninety five percent confidence interval using SPSS. When available, results were compared with project monitoring data and other program reports.

22

3. RESULTS

3.1 Context and Background Table 5: Project Intervention timeline

Date Project Activities Carried Out 2009

Jan-March

• Orientation and familiarization workshop • Establishing and Orientation for woreda food security task forces. • Beneficiary selection completed • One VSLA group with 20 members established

April-June

• Training on Improved Production techniques, financial record keeping and business skills provided to 125 cattle, 478 small ruminant (SR) and 862 cereal value chain participants.

• Livestock purchasing committee established • Purchasing of cattle and small ruminants began • Another 6 VSLA groups established.

July-Sept • Value chain inputs (credit) transferred for 124, 294 and 768 cattle, SR and cereal participants respectively. Oct-Dec • Solving Muslim credit challenge “ Loan Interest”

2011

Jan-March

• Conference conducted on Muslim credit issue on the program’s loan modality. • Final screening and selection of participants for livestock fattening • Construction of 5 deep boreholes to enhance rural water supply

April-June

• Final screening and selection of teff value chain (VC) participants • Training in improved production techniques, financial literacy and business skills for 192 cattle and1000 SR VC HHs, • Value chain input transfers for 28 HHs, in the cattle and 731 HHs, in the small ruminant value chain • Purchasing of Improved teff seeds • Organizing and establishing of VSLA groups • 15 groups engaged in small ruminant fattening involving 663 male and 457 female participants

July-Sept

• Training in improved production techniques, post harvesting, financial literacy and business skills for 1,733 teff VC HHs, • 260 quintals of improved teff seed distributed to HHs, • Assessment on teff germination rate carried out • Value chain input (livestock) transfer for 100 small ruminant VC participants • Installation of electric power to the constructed deep boreholes. • Financial literacy training for VSLA group members

Oct-Dec

• Entrepreneurship training for 100 female participants. • Selection for cattle and small ruminant participants. • Value chain input transfer for 70 small ruminant participants • Establishment of molasses distribution groups • Construction of stands and purchasing of molasses • Construction of livestock market center

2011

Jan-March

• Training in improved production techniques, financial literacy and business skills for 280 cattle VC participants. • Training in improved production techniques, financial literacy and business skills for 112 SR VC participants. • Financial literacy and business plan development training provided to 176 VSLA members. • Strengthening multipurpose cooperative in marketing and loan collection (for 291 members) • Value chain transfer for 100 cattle and 160 small ruminant participants • Training on how to use molasses, and mechanisms of distribution to PSNP + HHs, (38 participants).

April-June

• Value chain transfers (livestock) for 128 cattle VC participants • Multi Stakeholder Platform workshop on cereal and livestock VCs conducted • Refreshment training for multipurpose cooperative on marketing and loan collection for 200 members. • Final screening conducted for teff VC participants • Purchasing of improved teff

July-Sept

• Training in improved production techniques, post harvesting, financial literacy and business skills for 494 teff VC HHs, • Value chain input transfer (improved seeds) for 494 teff VC participants. • Follow up and assessment on germination rate of the improved teff seed. • Refreshment training (all topics) for 2000 PSNP plus participants from all value chains.

23

Table 6: Characteristics and background data on assessment participants

Household  (HH)  Background  and  Project  Participation  Stats   Sheep  2009  

(n=136)  Sheep  2011  (n=116)  

Cattle  2009  (n=82)  

Cattle  2011  (n=77)  

Total  number  currently  involved  in  cattle  value  chain   0   0   82  (100%)   77  (100%)  Total  number  currently  involved  in  sheep  value  chain   136  (100%)   116  (100%)   0   0  Total  number  currently  involved  in  teff  value  chain   0   8  (7%)   0   0  Total  number  currently  involved  in  VSLA   0   3  (3%)   0   10  (13%)  Number  of  HHs,  with  iron  sheet  roofing   22  (16%)   26  (22%)   6  (7%)   9  (12%)  Highest  level  of  education  HH  head  (average  grade)   1   0.9   1   0.6  Highest  level  of  education  other  HH  member  (average  grade)   4   5.8   4   4.4  Total  number  of  households  graduated  from  the  PSNP   8  (6%)   10  (9%)   1  (1.2%)   5  (6%)  Types  of  Shocks  Experienced  Drought/rain-­‐failure   135  (99%)   16  (14%)   82  (100%)   16  (21%)  Crop  pests   4  (3%)   16  (14%)   5  (6%)   12  (16%)  Livestock  disease   35  (26%)   58  (50%)   20  (24%)   26  (34%)  Illness/death  (HH  member)   44  (32%)   43  (37%)   23  (28%)   29  (38%)  

3.2 Income

3.2.1 Income Sources Figure 1: Changes in relative contributions of different income sources (Sheep)

24

Figure 2: Changes in relative contributions of different income sources (Cattle)

Notes on Figures 1-2 Method: Proportional piling using 100 counters P. Trade/IGA = Petty Trade/Income Generating Activities Figures 1-2 show changes in the relative contribution of different income sources since the project started. In terms of project impact the results show a significant increase in the relative contribution of income from livestock fattening for both intervention samples but no change in the contribution of this income source for the control group. This contribution is comparable with income contributions from other important sources such as income from teff sales and informal labor activities.

25

Table 7: Perceived changes in household income since 2009

Changes in income against a nominal baseline of 10 Category Mean score (95% CI) Percentage (increase) Sheep value chain (n=116) 13.7 (13.3, 14.1) 37% Control (n=105) 12.3 (11.9, 12.7) 23% Cattle value chain (n=77) 13.9 (13.1, 14.8) 39%

Data derived by scoring a total of 20 counters against a nominal baseline of 10 counters Table 8: Factors contributing to an increase in income

Factor   Mean  Score  (95%  CI)  Sheep  (n=101)   Control  (n=80)   Cattle  (n=65)  

PSNP   12.1  (9.7,  14.4)   13.6,  (11.2,  15.9)   10.6  (8.1,  13.2)  Fattening  Sales   8.4  (6.5,  10.3)   1.5  (0.1,  2.9)   10.5  (8.3,  12.6)  Teff  Sales   6.0  (3.8,  8.3)   5.5  (3.2,  7.8)   7.7  (4.8,  10.6)  Other  Reasons   23.5  (20.5,  26.5)   29.4  (26.4,  32.4)   20.5  (16.7,  24.3)  

3.3 Expenditure Figure 3: Relative Expenditure on Key Items (Sheep)

26

Figure 4: Relative Expenditure on Key Items (Cattle)

Notes on Figures 3-4: Agric Inputs = Agricultural/farming inputs Land/HI = Land rent and or Home Improvement (construction/maintenance) Livestock = Includes all livestock related expenses (animal health/fodder etc.) Social Oblig. = Social Obligations (weddings/funerals/contributions etc) Figures 3-4 show relative changes in key investments or expenditures since the project started. The results show no major changes in relative expenditure except for a decrease in food purchases for the sheep and control samples, and a decrease in medical expenses for the sheep sample. The results show an increase in relative expenditure on entertainment for the sheep sample, social obligations for the cattle and control samples, and an increase in clothes purchases for the control group.

27

3.4 Credit and Savings Table 9 Changes in the value of savings and loans

Savings  &  Loans  by  Source  2009  and  2011   Mean  Value  

(ETB)  (95%  CI)  

Lower   Upper  

Shee

p  

Savings  2008  (n=136)   192.4   27.8   356.9  Savings  2011  (n=116)   1873.2   1149.4   2596.9  MFI/DECSI  2009   406.8   191.9   621.6  MFI/DECSI  2011   560.3   293.8   826.8  Other/Private  2009   188.3   82.4   294.3  Other/Private  2011   101.1   28.4   173.8  Total  2009   598.0   365.0   831.1  Total  2011   2556.7   1763.6   3349.7  

Cattle  

Savings  2008  (n=82)   282.9   -­‐14.7   580.6  Savings  2011  (n=77)   2112.3   1143.9   3080.6  MFI/DECSI  2009   139.0   -­‐13.8   291.8  MFI/DECSI  2011   137.8   -­‐23.1   298.7  Other/Private  2009   182.0   30.9   333.2  Other/Private  2011   674.5   -­‐538.8   1887.9  Total  2009   321.0   111.2   530.9  Total  2011   2924.6   1321.8   4527.4  

Control  

Savings  2008  (n=130)   334.5   84.7   584.4  Savings  2011  (n=105)   628.4   367.9   888.9  MFI/DECSI  2009   544.9   332.5   757.2  MFI/DECSI  2011   342.1   160.3   523.9  Other/Private  2009   110.6   60.6   160.5  Other/Private  2011   93.5   14.0   173.0  Total  2009   655.4   445.8   865.0  Total  2011   1064.0   748.9   1379.1  

 

28

Figure 5: Utilization of savings and credit

Notes on Figure 5 Trade/IGA = Trade/Income Generating Activities Land/HI = Land rent/Home Improvement Livestock = Animal purchases only (does not include fodder/transportation/animal health expenditures) Agric Inputs = farming Inputs

3.5 Asset Levels and Changes

3.5.1 Land Holdings Table 10: Changes in land holdings

Sample   Land  Holdings  (tsimad)  Mean  Qty  (95%  CI)  2009   2010   2011  

Sheep     3.9  (3.5,  4.3)   3.9  (3.6,  4.3)   5.3  (4.6,  6.1)  Control     3.8  (3.3,  4.3)   3.9  (3.3,  4.4)   3.7  (3.1,  4.3)  Cattle     4.4  (3.9,  4.8)   4.7  (4.2,  5.2)   6.3  (5.5,  7.2)  

Sample Size Sheep 136 116 Control 130 105 Cattle 82 77

0 200 400 600 800 1000 1200 1400 1600 1800 2000

Livestock

Land/HI

Food

Clothes

Social Obligations

Health

Trade/IGA

Agric. Inputs

Education

HH Items

Other

Mean Expenditure (ETB)

Savings and Loan Utilization

Cattle n=77

Sheep n=116

29

3.5.2 Livestock holdings

Figure 6: Changes in livestock holdings (Sheep)

Figure 7: Changes in livestock holdings (Cattle)

See notes on Figures 6-7 next page

30

Figures 6-7 show changes in livestock holding since 2009. The results show a significant increase in calves and a significant decrease in cows for the sheep sample, and no significant changes in livestock holdings for the cattle sample since the project started. The results also show a significant decrease in oxen, cows, steers and heifers for the control sample (figures 7-8). This would suggest that project participants have been better able to maintain their livestock assets than non-project participants, although to some extent this might be attributed to project related livestock transfers.

31

3.5.3 Productive Assets/Tools Figure 8: Changes in productive assets (Sheep)

Figure 9: Changes in productive assets (Cattle)

Notes on Figures 8-9: Trad. Beehives = Traditional Beehives

32

3.5.4 Household Items Figure 10: Changes in household items (Sheep)

Figure 11: Changes in household items (Cattle)

Notes on Figures 10-11: Radio/CP = Radio/Cassette Player

33

3.6 Changes in Household Food Security Table 11 Food Security Calendar (n= 15 groups)

Time frame

Average Score Score/60 Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov

Before PSNP

26

2008 – 2009

15

2009 – 2010

35

Data derived from proportional piling exercises using a total of five counters per month (5 counters =Very Food Secure 0 counters =Food Insecure)

Table 12 Factors contributing to improvements in food security

Ranking  of  Factors  (n=15  groups)  1   2   3   4   5  

PSNP   Improved  rainfall   Irrigation   Livestock  credit    Improved  rainfall   PSNP   Livestock  credit  

   Livestock  credit   Improved  rainfall   PSNP      Improved  rainfall   PSNP   Livestock  credit      Livestock  credit   Labor   Irrigation   PSNP  

 Improved  rainfall   PSNP   Livestock  credit   Labor    Improved  rainfall   Improved  rainfall   PSNP   Livestock  credit    PSNP   Improved  rainfall   PSNP   Livestock  credit    Improved  rainfall   PSNP   Trade   Livestock  credit   Labor  

PSNP   Improved  rainfall   Trade   Labor   Livestock  credit  Improved  rainfall   PSNP   Livestock  credit   Trade  

 PSNP   Improved  rainfall   Livestock  credit   Trade   Labor  PSNP   Improved  rainfall   Labor   Trade   Livestock  Credit  Improved  rainfall   Livestock  credit   PSNP   Labor   Trade  Livestock  credit   Improved  rainfall   PSNP   Labor  

 Method: Simple Ranking (project factors highlighted in red)

34

3.7 Value Chain Production and Sales Table 13: Livestock value chain transfers and sales

Mean Qty/Value

Sheep (n=116) Cattle (n=77) Number of Project animals received 4.2 1 Number of fattening cycles 1 1 Number of animals fattened 3 1 Number of fattened animals sold 3 1 Income from the sale of fattened animals (ETB) 1,425 2,476 Figure 12: Utilization of income from sheep fattening

Notes on Figure 12 Agric Inputs = Farming Inputs Land/HI = Land/Home Improvement (land rent/construction/maintenance etc) Trade/IGA = Trade/Income Generating Activities Livestock = Animal Purchases only

0 50 100 150 200 250 300 350 400

Livestock

Food

Loans/debts

Clothes

Social Oblig

Trade/IGA

Land/HI

Education

Other

Agric. Inputs

Health

Mean Expenditure ETB

Income Utilization from Sheep Fattening (n=116)

35

Figure 13 Utilization of income from cattle fattening

Notes on Figure 12 Agric Inputs = Farming Inputs Land/HI = Land/Home Improvement (land rent/construction/maintenance etc) Trade/IGA = Trade/Income Generating Activities Livestock = Animal Purchases only

0 200 400 600 800 1000 1200 1400 1600

Livestock

Loans/debts

Food

Land/HI

Clothes

Health

Other

Agric. Inputs

Social Oblig

Mean Expenditure ETB

Income Utilization from Cattle Fattening (n=77)

36

3.8 Scoring of Project Benefits Table 14: Scoring of other project outcomes (Sheep)

    Sheep  Value  Chain  n=116   Score  1   Contributed  to  household  food  security   312  2   Enabled  us  to  protect  our  assets   263  3   Improved  our  resilience  to  drought  or  other  livelihood  shocks   258  4   Improved  our  skills/knowledge  on  livestock  fattening   251  5   Helped  us  recover  from  the  drought   201  6   Increased  our  income  from  livestock  (fattening)  sales   182  7   Improved  the  quality  of  fattened  animals  produced   175  8   Improved  our  status  in  the  community   173  9   Improved  our  relationship  with  neighbors  and  other  community  members   169  10   Increased  the  number  of  fattened  animals  produced   168  11   Enabled  us  to  accumulate  assets   167  12   Improved  our  knowledge  of  savings  and  finances   162  13   Increased  our  savings   150  14   Improved  our  access  to  inputs  for  livestock  fattening   144  15   Improved  our  business  skills   144  16   Improved  our  access  to  credit   127  17   Improved  our  financial  decision  making  in  the  household   112  18   Helped  us  cope  with  the  drought  in  2009   98  19   Improved  our  relationship  with  traders  and  or  the  private  sector   75  

Table 15: Scoring of other project outcomes (Cattle)

    Cattle  Value  Chain  n=77   Score  1   Contributed  to  household  food  security   204  2   Enabled  us  to  protect  our  assets   180  3   Improved  our  resilience  to  drought  or  other  livelihood  shocks   168  4   Improved  our  skills/knowledge  on  livestock  fattening   163  5   Increased  our  income  from  livestock  (fattening)  sales   129  6   Helped  us  recover  from  the  drought   128  7   Improved  our  relationship  with  neighbors  and  other  community  members   124  8   Enabled  us  to  accumulate  assets   120  9   Improved  our  status  in  the  community   118  10   Improved  our  knowledge  of  savings  and  finances   114  11   Increased  the  number  of  fattened  animals  produced   109  12   Improved  the  quality  of  fattened  animals  produced   109  13   Increased  our  savings   104  14   Improved  our  business  skills   100  15   Improved  our  access  to  credit   98  16   Improved  our  financial  decision  making  in  the  household   86  17   Improved  our  access  to  inputs  for  livestock  fattening   84  18   Helped  us  cope  with  the  drought  in  2009   78  19   Improved  our  relationship  with  traders  and  or  the  private  sector   53  

37

3.9 Strengths and Weaknesses of Program Interventions Table 16 Livestock Value Chain SWOT Analysis (n=27 groups)

Strengths   Weaknesses  

 • Increase  in  household  income  from  sale  of  fattened  animals  • Improved  household  livelihoods  • Increased  household  assets  • Alternative  source  of  income  (from  fattening)  • Timely  distribution  of  inputs  (livestock  transfers)  • Improved  livestock  management/production  (from  training)  • Business  literacy  and  skills  training  have  been  extremely  

beneficial  • The  purchasing  and  asset  delivery  system  has  improved  over  

time  • Insurance  procedures  have  improved  and  become  clearer  

over  time    

 • Poor  selection  of  livestock  (participants  not  included)  in  the  early  

stages  of  implementation  • Some  participants  expected  to  receive  livestock  but  received  small  

ruminants  instead  • Forced  loan  repayment  for  animals  that  died  • Loan  size  small  for  fattening  inputs1  • Some  participants  are  using  the  oxen  for  plowing  rather  than  

fattening  so  not  enough  fattened  cattle  are  being  supplied  to  attract  traders  to  the  area    

Opportunities   Threats/Challenges    

• Large  livestock  population  in  the  woreda  • Considerable  supply  of  feed  from  crop  residue  • High  demand  for  meat  (locally)  and  in  nearby  towns  • Road  construction  will  provide  better  access  to  markets  • Supplementary  feeds  can  be  sourced  from  Arbagele  

International  Livestock  Development  plc  (AILD)  • Good  stakeholder  support  system  (Office  of  Agriculture  and  

Development,  REST,  DECSI,  AILD,  and  Alamata  Agriculture  Research  Center)  

• Given  that  the  area  is  characterized  by  extensive  crop  production,  there  is  potential  to  use  crop  residues  to  overcome  livestock  feed  shortages  

 • Drought  and  resulting  feed  shortage  compels  people  to  sell  

livestock  before  they  are  fattened  or  increases  production  costs  as  supplementary  feed  is  required  

• Project  cattle  used  for  plowing  resulting  in  poor  body  fat  accumulation  

• No  insurance  against  the  death  of  project  animals2  • Shortage  of  veterinary  drugs  and  long  distances  to  animal  health  

centers  (for  some  tabias)  • Insufficient  number  of  uniformly  fattened  (or  a  critical  mass  of  high  

quality  animals)  to  attract  traders  and  slaughter  houses  to  the  area  • Livestock  losses  due  to  predators  • Still  some  reluctance  on  the  part  of  Muslims  to  purchase  livestock  

on  credit  • Loss  of  communal  grazing  as  land  is  being  sold  to  investors  for  

irrigated  crop  production    

1 Apparently this was only enforced when the participant could not provide any evidence of death or documentation of treatment for the animal

2 Insurance was provided for the death of project animals, but some participants were initially not aware of this, or not familiar with the preconditions for claiming this insurance. However, it appears as though this has now been addressed.

 

38

Table 17 Village Saving and Loan Association SWOT analysis (n=5 groups)

Strengths    • Provide  credit  at  better  interest  rates  than  private  

lenders  (5-­‐10%  as  opposed  to  40%)  • Poor  households  have  access  to  greater  capital  (through  

savings  and  credit)  than  they  would  normally  have  • Members  can  access  loans  when  needed  (timely)  to  pay  

for  healthcare,  medicine  or  farming  inputs  • Credit  can  be  used  to  help  members  invest  in  new  

livelihood  activities  and  diversify  their  income  sources  • Improved  saving  culture  

 

Weaknesses    • Low  working  capital,  and  small  loan  disbursements  

prevent  members  from  investing  in  individual  business  activities  

• Lack  of  business  opportunities  in  the  area  • Lack  of  business  skills  to  take  advantage  of  credit  • Religious  and  socio  economic  differences  between  

members  creates  conflict  over  modalities  such  as  contributions  and  interest  rates  

Opportunities    • Collective  (group)  investments  may  provide  members  

with  new  income  and  livelihood  options  • New  roads  being  built  in  the  woreda  will  bring  new  

business  opportunities  to  the  area  (VSLA  credit  will  allow  members  to  take  advantage  of  these)  

• PSNP  Plus  has  created  awareness  of  credit  and  savings,  and  an  increased  willingness  on  the  part  of  Muslim  community  members  to  take  out  loans  

Threats    • Frequent  drought  and  failed  harvests  results  in  loss  of  

income  and  undermines  members  ability  to  contribute  and  save  

• Disagreements  on  interest  rates  and  other  modalities  could  result  in  groups  disbanding  (particularly  where  religious  or  socio-­‐economic  differences  exist)  

• Disagreement  on  how  group  saving  should  be  utilized,  some  members  prefer  the  idea  of  collective  investments,  others  prefer  that  the  money  individually  shared  out      

39

4. DISCUSSION

4.1 Methodological Issues There were a number of methodological limitations to the assessment in Raya particularly around indicators and timing. Consistent with the project objectives, the study focused on measuring impact in terms of assets. Although physical and financial assets may be useful poverty measurement indicators, the LIS study findings to date, indicate that livestock represent the most meaningful asset in poor rural Ethiopian households. However, livestock die and reproduce, and they are continuously being sold and purchased, these dynamics make it difficult to measure impact in terms of livestock holdings, unless this were to be done over a protracted period. Seeing as assets are used as benchmarks for PSNP graduation, it is also possible that households may under report on these, as there may clearly be incentives for households to remain in the program. There were also issues around timing in that the baseline assessment took place a year after the project started. As such, a retrospective baseline approach was used and this may have resulted in a certain degree of recall bias. In terms of attribution, the classic scientific approach involves the use of a control population of non-project participants. This approach involves comparing a control group with a “treatment” or “intervention” population to determine statistical difference between the two groups, the assumption being that the control group has similar characteristics as the intervention group (Catley et al, 2008). In identifying a control group for the study in Raya Azebo, the assumption was made that PSNP participants would share similar characteristics as PSNP Plus households and so the same comparison group was used for both livestock value chains. However, it has been argued that the complexity, diversity and variance that defines and distinguishes people and communities is such that the conventional rigor associated with using a control makes no sense when applied to community development research (Chambers, 2008). This argument is acknowledged, as is the limitation that the approach used in no way captures the multiplicity of independent variables or characteristics that make two population groups similar or indeed truly comparable. As such, the characteristics used to define comparability were limited to the specific asset indicators being measured. Having said this, the results show no significant difference in baseline asset levels or relative income sources between the intervention and control groups suggesting that for these indicators, this was a fairly reliable control.

40

4.2 Income and Expenditure

4.2.1 Income Sources Formal employment opportunities are limited in Raya Azebo, and the rural economy of the woreda is almost exclusively characterized by agricultural production. Two surveys carried out in the Raya Valley in 1996 and 1997 indicated that 98-99 % of households derived most of their income from crop and livestock production (REST, 1997). Data generated by the LIU indicates that the sale of livestock and livestock products in a “good year” contributes an estimated 25-30% of household income for the poorest wealth categories and between 50-60 % for better-off households in Raya Azebo (DPPA, 2007). Farming activities are dependent on the Belg (February/March to April/May) and Kiremti (July to September/October) rains (REST, 1997; DPPA, 2007). Good soil fertility, and the availability of land make the area well suited to crop production, and in recent years there has been some investment in commercial farming in the area, characterized by mechanized and irrigated crop production (DPPA, 2007). Farmers living close to the edge of the valley also practice ‘flood irrigation’ by diverting temporary streams created by rainfall in the adjacent mountains. The main crops grown in the area are sorghum, teff and maize, with teff being produced as a cash crop as well as for own consumption (DPPA, 2007). A variety of pulses, such as lentils, chickpeas, horse beans, fruit including lemons, oranges and bananas, and vegetables such as onions, potatoes, tomatoes and cabbage are also grown in the woreda. However, the type and distribution of these crops varies according to elevation (REST, 1997). For example, fruit is typically found in lowland areas, and coffee and Khat (chat) are also grown as cash crops in the lowland areas. Crops such as barley and oil seeds are grown at higher elevations, and prickly pear can be found in the midland and lowland areas (REST, 1997). According to LIU data, in a “good year”, roughly 20-30% of household income for the two poorest categories is derived from crop sales (DPPA, 2007). Aside from income derived from crops and livestock, there are limited economic or formal employment opportunities in the area. As such, in 2008-2009 the PSNP represented the single most important source of income for assessment participants, with the results indicating that almost one third of their income comes from this source (figures 1-2). Similarly, LIU data estimates that roughly 30-40% of income for the two poorest categories comes from the safety net program (DPPA, 2007). The results from the final assessment show a decrease in dependency on the PSNP for cattle value chain participants relative to the control sample (figure 2). Some informal seasonal agricultural work is available for poor households, who will work on the farms of better-off households during the harvesting and weeding periods (DPPA, 2007). Other economic activities include, petty trading (including khat and grain trading), firewood, fodder and stone selling, clay making and weaving. The results show a significant increase in the relative contribution of income from livestock fattening for both intervention samples but no change in the contribution of this income source for the control group (figures 1-2). This contribution is comparable with income

41

contributions from other important sources such as income from teff sales and informal labor activities (figures 1-2). Consistent with these findings, an increase in income from livestock sales was scored as the fifth and sixth most important project benefit by participants in the two value chain samples (tables 17-18). One of the key objectives of the livestock value chain interventions is to diversify peoples’ income and reduce their dependency on crop production and sales (SNV, 2010). Consistent with this objective, the results indicate that the fattening value chains have been effective in promoting livelihoods diversification by providing participants with an alternative source of household income.

4.2.2 Actual Income and Relative Expenditure Assessment participants across all three categories have experienced an overall increase in income since the project started with project participants seeing a greater increase in comparison to the control group. The results show that both intervention (fattening) samples experienced a 37-39% increase in income, representing a significant increase in comparison to the control sample who experienced a 23% increase (table 9). In absolute terms, 84% of the cattle sample and 87% of the sheep sample experienced an increase in income in comparison to 76% of the control group (table 10). The results indicate that the livestock value chain interventions have been an important contributing factor to this increase in income, with both intervention samples assigning a comparable score for fattening, teff sales, and the PSNP relative to other factors (table 10). The results show no major changes in relative expenditure except for a decrease in food purchases for the sheep and control samples, and a decrease in medical expenses for the sheep sample (figures 3-4). The results show an increase in relative expenditure on entertainment for the sheep sample, social obligations for the cattle and control samples, and an increase in clothes purchases for the control group (figures 3-4).

4.3 Credit and Savings A number of formal and informal microfinance options are available to rural communities in Raya Azebo. Dedebit Credit and Saving Institute is the main provider of formal financial products, offering interest-based loans to rural households. Formal agricultural credit is also provided through the Other Food Security Programs (OFSP). These include improved seeds and livestock loans provided by the woreda Agriculture and Rural Development offices, REST, and farmer’s cooperatives. Informal financial products are also available from private sources and traditional saving and lending groups such as Iqqub and Iddir. Focus group participants indicated that these primarily provide an insurance mechanism, whereby friends and neighbors will raise funds to cover unexpected shocks or medical emergencies, and to pay for burial services. Although uncommon, credit is sometimes provided to members of these groups. People also borrow money directly from individual neighbors or better-off community members. Typically, interest on these loans is high ranging from 10-100%. These can either be paid back in cash or crops after the harvest. However, due to the high risk of rain and crop failure, people generally prefer to save rather than borrow, and convert their savings into livestock assets as a form of insurance against idiosyncratic and covariant livelihoods shocks. Since the project started, there has been a significant increase in the value of savings and loans for project participants and no significant increase for control group participants (table

42

9). One explanation for this might be that the project has promoted a savings culture amongst project participants, and typically this would have been facilitated through the informal microfinance component of the project. However, this would appear unlikely given that this component was limited to a pilot of seven VSLA groups, and there was very little overlap between VSLA participants and the study samples (table 6). Alternatively, it is possible that project participants are financially in a better position to save as a result of their involvement in the project. If so, this may be due to the extra income earned through livestock fattening. The results show no significant change in the value of loans accessed since the project started (table 9). However, this does not necessarily imply that there has been no improvement in access to microfinance as involvement in the livestock credit component of the value chain intervention may have excluded participants from receiving further loans. Furthermore the project in Raya has been innovative in developing new and culturally appropriate loan products in the area (see section 1.3.2), and participants scored livestock credit as an important factor contributing towards improved household food security (table 12).

4.4 Changes in Assets Assessment participants indicated that the project has helped people protect their assets, this being scored as the second most important project benefit by both the cattle and small ruminant value chain samples (tables 14-15). The results also indicate that project participants have experienced some positive changes in certain physical and financial assets since the project started, with asset accumulation also being scored as a somewhat important project benefit (tables 14-15).

4.4.1 Land Holdings The findings from the mid term assessment indicate that land holdings are calculated in terms of people’s ability to utilize their land effectively. This factor is largely determined by household labor capacity and the ownership of draft animals (Burns and Bogale, 2010). As such, land utilization might be seen as a useful proxy for livestock ownership and household productive capacity. When viewed from this perspective, the results are encouraging as they indicate that there has been a significant increase in land holdings (utilization) for both intervention samples since the project started (table 10). In comparison, there has been no significant increase in land holdings for the control sample (table 10). The two most frequently mentioned reasons for an increase in land holdings across all three samples were access to labor and access to draft animals (Annex II). These results would suggest that PSNP Plus participants have experienced a significant increase in productive capacity since the project started.

4.4.2 Livestock Holdings Assessment participants identified livestock as the key determinant of wealth in the Raya Valley with oxen being the single most important wealth indicator (Annex I). The results show a significant increase in calves and a significant decrease in cows for the sheep sample, and no significant changes in livestock holdings for the cattle sample since the project started (figures 7-8). The results also show a significant decrease in oxen, cows, steers and heifers for the control sample (figures 7-8). This would suggest that project participants have been

43

better able to maintain their livestock assets than non-project participants, although to some extent this might be attributed to project related livestock transfers. Nonetheless, the results from the mid term assessment showed a significant decrease in oxen and heifers for both intervention samples, and a significant decrease in cows for the sheep sample (Burns and Bogale, 2010). This was largely attributed to stress sales and livestock mortality associated with the drought in 2009 (Burns and Bogale, 2010). As such, the results indicate that project participants have been better able to protect and maintain their livestock assets relative to the control group.

4.4.3 Productive Assets and Household Items The results show that there have been some significant changes in productive assets and household items since the project started. The final results show a significant increase in axes’ and hoe’s for the sheep sample, and pickaxes for the cattle sample (figures 9-10). Aside from pickaxes’ for the cattle sample, the results show no significant difference in these tool holdings between the intervention and control group (figures 9-10). However, the final results show the cattle sample having significantly more ploughs than the control, a difference not evident during the baseline (figure 10). Although somewhat inconclusive, the results suggest that the intervention samples have been better able to increase their productive assets relative to non-project participants. In terms of household items, the results show a significant increase in cups and mats for both intervention samples relative to both the baseline and the control group (figures 11-12). The results also show a significant increase in Jeri cans for the cattle sample, again relative to both the baseline and the control (figure 12).

4.5 Changes in Food Security The results from the focus group discussions show a considerable improvement in both temporal and absolute food security since the PSNP was launched (table 11). Prior to the PSNP households would typically face a three months food deficit from July to September (table 11). This deficit now appears to have been eliminated. Even during the drought (2008-2009) no temporal food deficit was reported even though absolute food security was lower than pre-PSNP estimates suggesting that the PSNP has been effective in addressing chronic food insecurity even during drought periods. Consistent with this, along with improved rainfall, the PSNP was ranked as one of the most important factors contributing towards household food security (table 12). However, participants also ranked livestock credit as an important contributing factor with three of the fifteen focus groups assessed ranking this as the most important contributing factor (table 12). The results from the household component also support these findings in that the highest scoring project benefit for both value chains was the projects contribution towards household food security (tables 14-15).

4.6 Utilization of Project Derived Income and Credit

4.6.1 Savings and Loan Utilization The results show that (mean) household savings and loans combined for 2010-2011 was estimated at 2,535 Ethiopian birr for the sheep sample, and 1,763 birr for the cattle sample (table 9). This money was invested in a variety of livelihoods activities with the most

44

important expenditures including land, livestock, food, clothes and education (figure 6). There were however some differences in the importance of these between the samples. For example, livestock investments were the most important expenditure for the cattle sample, whereas food purchases were the most important for the sheep sample (figure 6). These results suggest that savings and loans may have contributed towards asset accumulation and improved food security for project participants.

4.6.2 Utilization of Income from Livestock Fattening By the time of the final assessment, mean income from sheep fattening was estimated at 1,425 Ethiopian birr, and the mean income from the sale of fattened cattle was estimated at 2,476 birr (table 13). This income was spent on a variety of livelihoods activities with livestock investments being the most important expenditure for both value chain samples (figures 13-14). For example, sheep value chain participants’ reinvested 346 birr (mean expenditure) in livestock, and cattle value chain participants’ reinvested 1,432 birr (figures 13-14). Over time, it might be expected that livestock assets would continue to increase as a result of these investments. Other important investments included food purchases, and the repayment of loans and debts (figures 13-14).

CONCLUSION   The results from this assessment appear to validate the programs causal model by providing some evidence that the combination of microfinance and value chain interventions leads to asset accumulation and graduation out of chronic food insecurity. Since the project started, value chain participants have experienced a significant increase in income and savings, and considerable improvements in food security. They have also seen a significant increase in land holdings, which may represent an increase in household productive capacity, and a significant increase in certain types of productive and household assets. The project has also evidently helped people protect and maintain their livestock assets. Furthermore, meaningful amounts of project derived income savings and credit are being invested in various livelihoods assets that could potentially translate into further production income and other livelihoods benefits over time. The project value chains have also helped to strengthen and possibly even diversify participants’ income sources. As positive as these results are, what is particularly encouraging is that these impacts occurred against the backdrop of a drought that affected the study area during the first year of project implementation. The drought resulted in crop and income losses, as well as food and livestock feed shortages and moderate levels of livestock mortality. People also expanded on a number of coping strategies, such as the sale of livestock in order to cope with the drought. The results therefore show remarkable recovery and resilience amongst project participants. In light of these results, there are a number of lessons that can be drawn from this case study that may be applicable to the design and implementation of future livestock value chains under the HABP and Graduation with Resilience to Achieve Sustainable Graduation (GRAD) initiatives.

45

Firstly the project appears to have been well designed and appropriate for this particular context. As with all value chains, a thorough contextual analysis is essential in determining the viability of the intervention and in Raya, the findings from both the mid term and final assessment suggest that the project was designed with a comprehensive understand of the physical context and capacity of the intended participants. Much of this had to do with REST’s long-term presence on the ground and ability to retain quality field staff with decades of experience working with the same communities. In the absence of this capacity, future programs such as GRAD and HAPB should invest in a comprehensive contextual analysis that goes beyond just a value chain assessment. This analysis should go beyond simply assessing the viability of a given value chain, but needs to be more nuanced to capture the risks associated with a particular value chain, and the capacities of potential participants to full utilize or benefit from a particular value chain or credit package. The project was also well implemented with no major delays in the delivery of activities and inputs, for example 69% of targeted asset transfers for the sheep sample, and 100% of asset transfers for the cattle sample was achieved during the first year of implementation (REST, 2009; CARE, 2009). This can largely be credited to good coordination and planning, although arguably in the context of highland Ethiopia, livestock value chains are fairly straight forward to implement in terms of input procurement, training and marketing. Granted, a project should not be chosen for convenience alone, however if a given intervention can achieve the desired results as appears to be the case in Raya, then practicability should be weighed into the consideration of livestock value chains under both HABP and GRAD. It would also appear from this and evidence from other LIS case studies, that livestock production is more resilient to drought and weather related shocks than rain dependent crop value chains, and the fattening value chains appear to have been less affected by the drought in 2009 than the cereal and white pea bean value chains being implemented under the PSNP Plus program. This evidence would support the inclusion of livestock fattening value chains in drought-affected woredas’ under HABP, GRAD and other food security programs. The income and livelihoods benefits from fattening are also more immediate than those from crop value chains given the shorter production cycle, particularly for sheep fattening. As such, where the production and marketing potential exists, livestock fattening would appear to be an appropriate intervention in drought prone areas where short to medium term impact is desired. Having said this, the longer- term impacts from these interventions is still unknown and it is unclear to what extent they might be scaled up and sustained in a specific area. It should be noted that the process of fattening described by participants would probably not be classified as scientific livestock fattening given the feed shortage constraints in the area. The process described was more akin to livestock speculation and trading involving elements of natural or less scientific fattening. However, the flexibility for people to tailor this type of intervention to different needs is a compelling reason for the inclusion of livestock credit under HABP and GRAD. For example, households with the resources to purchase all the inputs for scientific fattening can effectively participate in real value addition and benefit from a greater return on their investments. Poorer households can use the credit to purchase livestock to increase their assets through livestock rearing, and households whose cash needs are more immediate can speculate and trade livestock without investing in all the inputs required for

46

fattening. As such, given its flexibility, livestock credit would appear to be able to effectively target different household typologies with different needs and capacities. In Raya, the potential for expanding livestock marketing has been assessed, and favorable policies and agro-ecology, good livestock breeds (Raya, Harmo and Adel) and a high local demand for meat exists (SNV, 2010). However, the shortage of quality pasture and water, along with poor animal livestock management practices and livestock disease have been identified as major constraints to livestock marketing in the area (REST, 1997; TARI/AARC 2010; SNV, 2010). These constraints are no doubt applicable to many parts of the country and clearly have implications on sustainability and scaling up under future initiatives. The lack of quality animal health services is likely to significantly undermine any livestock value chain intervention. Although reported livestock disease and mortality in Raya was fairly low, in another LIS study area animal disease outbreaks in concert with a shortage of animal health workers resulted in considerable losses of project animals. Under both GRAD and HAPB, where livestock value chain interventions are being considered, an assessment of the local animal health services in terms of availability, quality and accessibility should be included as part of the initial value chain assessment. The availability of quality feed also needs be considered when identifying potential livestock value chain intervention areas. For example, feed shortage has been ranked as the number one constraint to livestock production and marketing in Raya (TARI/AARC, 2010). Assessment participants also identified the issue of livestock feed as a major project challenge (table 16). Consistent with this, another study suggests that the livestock population in the Raya Valley, exceeds its carrying capacity, in that pasture is poor due to overgrazing, and that the valley experiences an estimated net livestock feed shortage of four and a half months each year (REST, 1997). Of particular concern is the likelihood that the availability of pasture and the quality of pasture will decrease over time as a result of agricultural expansion, the sale of communal grazing land to investors and continued overgrazing. Given these constraints, an increase in the off-take of livestock through increased sales is indeed desirable, and livestock fattening interventions could potentially provide the mechanism for this to happen. However, for such interventions to be economically and environmentally sustainable will require investments in the local animal feed industry and a focus on quality production and sales as opposed to quantity oriented livestock production. Similarly, potential livestock value chain intervention woredas’ under HABP and GRAD are likely to face similar constraints in terms of feed availability and carrying capacity. Such constrain need to be carefully considered as they will determine the extent to which these programs can be scaled up and sustained.

47

References and Further Reading  Burns, J., and Bogale, S. (2010) Linking Poor Rural Households to Microfinance and Markets in Ethiopia; Baseline and Mid-Term Assessment of the PSNP Plus project in Raya Azebo, Feinstein International Center, Medford, MA Borchgrevink, A., Helle-Valle, J., and Wodehanna, T. (2003) Credible Credit Impact Study of the Dedebit Credit and Saving Institution (DECSI), Tigray, Ethiopia. Norwegian Institution of International Affairs and Norsk Borchgrevink, A., Wodehanna, T., Ageba, G., and Teshome, W. (2005) Marginalized groups, credit and empowerment: The case of Dedebit Credit and Saving Institution (DECSI) of Tigray, Ethiopia. Occasional Paper No 14. AEMFI. Addis Ababa CARE (2010) PSNP Plus Project, Linking Poor Rural Households to Microfinance and Markets: Seventh Quarterly Report April- June 2010, Addis Ababa. CARE (2010) PSNP Plus Project, Linking Poor Rural Households to Microfinance and Markets: Sixth Quarterly Report January-March 2010, Addis Ababa. CARE (2009) PSNP Plus Project, Linking Poor Rural Households to Microfinance and Markets: Year One Annual Report and 4th Quarter Programmatic Report; July 2009 to September 2009. Addis Ababa. Catley, A. Burns, J, Abebe, D, and Omeno, W. 2008: Participatory Impact Assessment: A Guide for Practitioners, Feinstein International Center, Medford, MA 2008 Chambers, R. (2008) Revolutions in Development Inquiry: Institute of Development Studies, 2008, London, Earthscan. Devereux, S., Sabates-Wheeler, R., Tefera, M and Taye, H. (2006) Ethiopia’s Productive Safety Net Programme (PSNP), Trends in PSNP Transfers Within Targeted Households, Final Report. IDS: Brighton UK, Indak, International, Addis Ababa 2006. DPPA (2007) Tigray Livelihood Zone Reports: Raya Azebo Woreda, Southern Administrative Zone. Raya Valley Sorghum and Teff Livelihood Zone: Report from the Livelihoods Information Unit, Disaster Preparedness and Prevention Agency (DPPA) Addis Ababa, February 2007. Getahun, M (2010) “The Plus” In Getahun, M (ed) PSNP Plus newsletter: Volume 1. Issue 2, Addis Ababa, October 2010. Gilligan, D., Hodinott, J., Kumar, N., Taffasse, A, S., Dejene, S., Gezahegn, F., and Yohannes, Y. (2009) Ethiopia Food Security Program: Report on 2008 Survey. International Food Policy Research Institute (IFPRI). Washington, D.C. Micro Development Training and Consultancy Services (2010) Final Report – Financial Product Development and Linkage Mechanism – PSNP Plus, Addis Ababa. Ministry of Agriculture and Rural Development (MoARD, 2009), Food Security Programme 2010-2014: Household Asset Building. Ministry of Agriculture and Rural Development, Addis Ababa, August 2009 Ministry of Agriculture and Rural Development (2007), Productive Safety net Programme: Graduation

48

Guidance Note, Food Security Co-ordination Bureau, Ministry of Agriculture and Rural Development, Addis Ababa, December, 2007 Ministry of Agriculture and Rural Development (2006) Productive Safety Net Programme: Programme Implementation Manual. Addis Ababa, Ministry of Agriculture and Rural Development, Government of the Federal Democratic Republic of Ethiopia. Pankhurst (2009) Rethinking Safetynets and Household Vulnerability in Ethiopia: Implications of Household Cycles, Types and Shocks. Paper presented at the World Conference of Humanitarian Studies, Groningen, Netherlands. 4-7 February 2009 PSNP Plus Project Proposal (2008) Bringing Ultimate Yields through Integrated Networks (BUY IN): Linking Poor Rural Households to Microfinance and Markets REST (2009) PSNP Plus: Linking Poor Rural Households to Microfinance and Markets in Ethiopia: Annual Intermediate Results and Outcome Report. December 2009. Relief Society of Tigray (REST). Makelle, Ethiopia. REST (1997) Feasibility Report of the Raya Valley Development Study Project: Volume 1. The Federal Democratic Republic of Ethiopia, Tigray National Regional Government, The Relief Society of Tigray (REST). Alamata, August 1997. Sabates- Wheeler, R., and Devereux, S. (2010) Cash transfers and high food prices: Explaining outcomes on Ethiopia’s Productive Safety Net Programme. Food Policy article in press (2010) Sharp, K., Brown, T, and Teshome, A. (2006) Targeting Ethiopia’s Productive Safety Net Programme (PSNP) Overseas Development Institute (ODI) August 2006.

Slater, R., Adhley, S., Tefera, M., Butta, M, and Esubalew. (2006). PSNP Policy, Programmes and Institutional Linkages, Final Report. SNV (2010) PSNP Plus Livestock Sheep and Goats Fattening Value Chain Analysis Report: SNV, Ethiopia. Addis Ababa, Ethiopia, March 2010. TARI/AARC (2010) Problem Appraisal and Prioritization of Southern Zone Tigray (Alamata, Raya Azebo, Endam-Mokoni and Emba Alaje) Woreda’s. Tigray Agricultural Research Institute (TARI), Alamata Agricultural Research Institute (AARC), Alamata, January 2010. USAID (2008) Linking Poor Rural Households to Microfinance and Markets; RFA No: 663-A-08-015. USAID, Ethiopia, Addis Ababa, March 2008.

49

ANNEXES

Annex  I  Community  wealth  indicators  (n=27  Groups)   Wealth  Indicator   Better-­‐Off   Medium   Poor  Percentage  of  the  Population   15% 25% 60% Oxen 4 2 0-1 Cows 5 3 0-1 Calves 2 1 0-1 Steers 2 1 0-1 Heifers 2 1 0-1 Small Ruminants 11 6 3 Donkeys 2 1 0-1 Camels 9 3 0 Chickens 3 2 4 Traditional beehive 2 1 0-1 Overall landholding (tsimad) 7 7 7 Cultivated land 7 7 6 Beds 1 1 0 Mattresses 1 1 0 Mats 1 1 0 Lanterns 1 1 1 Barrel 1 0 0 Radio/Cassette Player 1 1 0 Corrugated iron sheet house roofing (No of sheets) 34 16 1 Number of rooms in the house 1 1 0 Food Security from own production (months) 12 11 6 Food Security from other sources (months) 0 1 6 Critical food deficit (months) 0 0 5 Engaged in labor activities for income NO NO YES Engaged in labor for farming oxen NO NO YES Other specify (tends others animals) NO NO YES Source: Burns & Bogale 2010

50

Annex  II  Ranking  of  Reasons  for  an  Increase  in  Land  Holdings  

Sheep  VC  (n=32)   Cattle  VC  n  =29  1st   2nd   3rd   1st   2nd   3rd  

Labor access Increased income

Availability of ox Labor access

Availability of Ox Labor access Labor access land rent Availability of Ox Land access     Availability of ox Labor access Availability of ox Land access Land redistribution

Increased income

Availability of ox Availability of ox Labor access

Labor access Availability of ox Labor access Availability of ox Labor access Availability of ox

Availability of ox Availability of ox Labor access Grain sales

Availability of ox Labor access Availability of ox Availability of ox Labor access Availability of ox Labor access Labor access Labor access Availability of ox Labor access Availability of ox Labor access Availability of ox Labor access Availability of ox Labor access Availability of ox Labor access Availability of ox Labor access Availability of ox Availability of ox Labor access Labor access Improved income Availability of ox Labor access Labor access         Availability of ox Land rent Availability of ox Availability of ox Labor access Availability of ox Labor access Availability of ox Labor access     Availability of ox Labor access Labor access     Labor access Availability of ox Availability of ox Labor access Availability of ox Labor access Labor access Availability of ox Labor access Labor access Availability of ox Labor access Availability of ox Labor access Labor access Availability of ox Availability of ox Availability of ox Labor access Availability of ox Labor access Availability of ox Labor access Labor access Availability of ox Labor access Labor access Availability of Ox            Availability of ox Labor access            

51

Annex  III  Household  Checklist    

Household Component Checklist FINAL IMPACT ASSESSMENT PSNP PLUS LIS/  Raya  Azebo

HH Code _______  

Name  of  Interviewer________________________________  Day:________  Month:  _________Year:    2011  

Woreda Raya Azebo Tabia Kushet/Village (cluster) 1. Household and Project Background Information

Household Code # Circle the appropriate boxes Gender HH. size Name of respondent M F Household roofing material (circle) Grass Corrugated Sheeting Project Activities that household members are involved

Cattle Small ruminant Cereals VSLA

Education/grade of Household Head Maximum education/grade of any household member Types of HH shocks experienced in the past 12 months (Yes/No) Rain-failure/drought Crop pest/disease Livestock disease/death Human illness/death Has your household graduated from the PSNP since the project started? YES NO

1b. If the household has graduated from the PSNP, ask the participant to list the most important factors that enabled them to graduate from the program and rank these in order of importance (5 most important reasons)

Reasons contributing to PSNP graduation RANK 1st 2nd 3rd 4th 5th

2. Savings and Loan Information

Circle the appropriate boxes How much money has your household managed to save since in the

past year? (including interest earned from VSLA contributions) Savings ETB Share ETB

A B Has your household taken out a loan in the past years? YES NO C (If Yes) – How much money did you borrow? ETB D Who did you borrow the money from? VSLA MFI Other E How much did you borrow (ETB) by source F Have you repaid the loan and interest? (Y)=Yes (N) = No

52

2b. How did your household spend your savings and loans? (in the past year only)

Ask the respondent to specify the amount by source (include VSLA savings and dividends for this source) Savings & Loan Utilization Amount ETB

VSLA MFI Other 1 Food & household consumables 2 Medical costs 3 Education/schooling (fees/uniforms/rent) 4 Land rent/property or home improvements (corrugated roofing etc.) 5 Purchase livestock or poultry 6 Invested in petty trade/retail or other business 7 Farming inputs (animal treatment/seeds/fertilizers/pesticides/tools) 8 Social obligations/ceremonies (weddings/funerals other contributions) 9 Clothes, shoes, blankets

10 Household items 11 Other (specify) (transport costs, tax, loan, entertainment etc.)

3. Asset Inventory 3a. How much of the following do you own - or did you rent or cultivate (this year)? Land 3b. Ask the respondent if they have increased the amount of land cultivated or number of trees since the project started. If the answer is yes, ask them to rank the 3 most important reasons for this increase. Reason Rank 1 2 3

3c. Using the following table, ask the participants how many of each of the following assets they currently own? (For livestock, do not include any animals that you are looking after but belong to someone else)

Livestock Type Sold

No Productive Asset Type No Household Item Type No

Oxen/bulls Plough and its accessories

Bed

Cows Sickle Mattresses Steers Pick Axe Mats Heifers Axe Chairs Calves Hoe Cupboards Sheep Spade Jericans Goats Grain mill (hand) Pots/Pans Donkeys Grain Mill (diesel) Cups Poultry Traditional beehive Lanterns Camel Modern beehives Tables Bee colony Radio or cassette player Charcoal stove £ Stress sale Kerosene stove £ Normal sale Mobile phone Barrel

53

3d. Ask the participant if there has been an increase in their total livestock holdings, or in the value of their livestock

holdings since the project started. If the answer is yes ask them to score the factors contributing to this increase against the following indicators: Method proportional piling using 100 counters (if the factor did not contribute – put zero) Contributing factors Score Credit from MFI (ACSI) Credit from other source Purchased with VSLA loan or income from VSLA investments* Purchased with income from Teff sales Purchased with income from other crop sale Purchased with income from fattening Purchased with income from the sale of livestock or livestock products Purchased with PSNP income Purchased with income from petty trade and any other source (IGAs)* Purchased with income from labor We were given this asset Livestock reproduced/matured Other reason Total 100

*Include all other income sources trading, handcrafts, brewing etc.

3e. Now repeat the same exercise for any reported increase in Productive assets (tools) or Household items Method proportional piling using 100 counters Contributing factors Score Credit from MFI (ACSI) Credit from other source Purchased with VSLA loan or income from VSLA investments* Purchased with income from Teff sales Purchased with income from other crop sale Purchased with income from fattening Purchased with income from the sale of livestock or livestock products Purchased with PSNP income Purchased with income from petty trade and any other source (IGAs)* Purchased with income from labor We were given this asset Livestock reproduced/matured Other reason Total 100

*Include all other income sources trading, handcrafts, brewing etc. 4. Income from FATTENING UTILIZATION What quantity of the following products from your own (farm) production did you sell from the last harvest? 1 2 3 4 Commodity Produced/finished Sold Income A Cattle fattening No. No. ETB B Shoat fattening No. No. ETB C Cereals/Teff Kg Kg ETB

54

4b. Take the amount (ETB) from A4 and B4 (previous page) and put it in the Total cell in the following table Check each of the items that apply and ask the respondent to specify the amount

Income Utilization

Amount ETB Cattle Shoat

1 Food & household consumables 2 Medical costs 3 Education/schooling (fees/uniforms/rent/transport) 4 Land renting/ purchase property or home improvements (corrugated roofing etc) 5 Purchase livestock or poultry 6 Invested in petty trade/ other trade retail, business or IGA 7 Farming & livestock inputs (animal health treatments/seeds/fertilizers/pesticides/tools,

feeds etc.)

8 Social obligations/ceremonies (weddings/funerals/iddir -other contributions) 9 Clothes, shoes, blankets

10 Other-specify (taxes, loan, transports, entertainment, HH items) Total A4= B4=

5. TOTAL HOUSEHOLD EXPENDITURE 5a. Last year - how much of your household income did you spend on the following items? (if nothing put zero)

Expense/Investment Score 1 Food & household consumables 2 Medical costs 3 Education/schooling (fees/uniforms/rent/transport) 4 Land renting/ purchase property or home improvements (corrugated roofing etc.) 5 Livestock or poultry purchases 6 Invested in petty trade/ other trade retail, business or IGA 7 Farming & livestock inputs (animal health treatments/seeds/fertilizers/pesticides/tools) 8 Social obligations/ceremonies (weddings/funerals/iddir -other contributions) 9 Clothes, shoes and blankets

10 Entertainment 11 Others (specify) (transport costs, taxes, loans etc.) TOTAL 100

55

6. INCOME

What proportion of your annual household income comes from the following sources? Method: Proportional Piling with 100 counters - (if nothing put zero)

6b Income Changes Method: Scoring against a nominal baseline of 10 counters

Before (counters)

Now (counters)

For this exercise, use 10 counters to represent the participants’ total household income before the project started. Now ask the respondent to compare this with their current income - by either adding or taking away counters to show an increase or decrease in total household income.

10

6c. If there has been an increase in income – ask the participant to score the project, non-project and

PSNP factors contributing to this improvement: Method (proportional piling with 50 counters)

Reason SCORE PSNP Fattening value chain Cereal (teff) value chain VSLA Other TOTAL 50

6d. Now ask them to specify and rank the other reasons in order of importance: Method: Simple Ranking (top 5 reasons only) Other reasons for changes in household income Rank 1st 2nd 3rd 4th 5th

Income Source Score 1 Teff sales from own farm production 2 Other crop sales (sorghum, maize vegetables, chat etc.) from own farm production 3 Income from livestock fattening 4 Income from other livestock & poultry production (meat/milk/eggs, trading etc.) 5 Petty Trade/retail and other IGA 6 PSNP work 7 Other labor/employment 8 Firewood/sand stone sale/fodder sales etc. 9 Handicrafts & cottage industry (knitting/ local drinks/ basket weaving etc.) 10 Other (specify) TOTAL 100

56

7. Other Project Benefits 7a. On a scale of zero to three - To what extent has the project:

(0=not at all; 1= a little; 2=medium; 3= plenty) Method (either simple ranking or proportional scoring depending on literacy level) SCORE Enabled you to accumulate assets? Enabled you to protect your assets? Contributed to household food security? Helped you cope with the drought in 2009? Helped you recover from the drought? Improved your business skills? Improved your knowledge of savings and finances? Increased your savings? Improved your access to credit? Improved your access to inputs for fattening? Improved your skills/knowledge on fattening? Helped you increase the number of finished animals? (quantity) Helped you finish/fatten the animals very well? (quality ) Increased your income from fattening? Improved your relationship with traders and or the private sector? Improved your financial decision making in the household Improved your relationship with neighbors and other community members Improved your status in the community Improved your resilience to drought or other livelihood shocks b. Ask the participants to identify any other important project benefits and give them a score of 1-5 (Up to 5 other benefits only) Other project benefits Score We would like to thank you for your time. Do you have any questions that you would like to ask us, or is there anything else you would like to tell us about the project, and how it might be improved?

Note: Consolidated version of checklist – translated summarized and formatted for the report)

57