This publication was produced for the United States Agency for International Development (USAID) by Social Impact, Inc. in collaboration with Ms. Rachel Bahn and Mr. Bashar Berro. Note: Limited redactions have been made to this version of the report in accordance with the principled exceptions to the presumption in favor of openness established in OMB Bulletin 12-01, “Guidance on Collection of U.S. Foreign Assistance Data.” LEBANON INVESTMENT IN QUALITY (LINQ) MID-TERM PERFORMANCE EVALUATION FINAL REPORT September 2020 Women’s cooperative members benefit from a training on food processing through LINQ. Source: USAID Photo Archive LEBANON INVESTMENT IN QUALITY (LINQ) MID-TERM PERFORMANCE EVALUATION FINAL REPORT September 2020 Prepared by Performance Management and Support Program for Lebanon II (PMSPL II) Contracted under AID-268-C-15-00001 This document is not available in print. Documents of this nature are made available to the public through the Development Experience Clearinghouse (DEC) repository (dec.usaid.gov). For additional information, please contact: Social Impact, Inc. Corporate Office 2300 Clarendon Boulevard Suite 10300 Arlington, VA, 22201 Tel: (703) 465-1884 Fax: (703) 465-1888 info@socialimpact.com Or Social Impact, Inc. Arz Street Librex Bldg. Bloc B – 3rd Floor Zalka, Metn, Lebanon Tel: +961-1-879260 DISCLAIMER The authors’ views expressed in this publication do not necessarily reflect the views of the United States Agency for International Development or the United States Government. i | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT TABLE OF CONTENTS ABSTRACT........................................................................................................................................................................ iii ACKNOWLEDGEMENTS ............................................................................................................................................. iv ACRONYMS....................................................................................................................................................................... v EXECUTIVE SUMMARY..................................................................................................................................................1 EVALUATION BACKGROUND..................................................................................................................................7 ACTIVITY BACKGROUND......................................................................................................................................7 ACTIVITY OBJECTIVES..............................................................................................................................................7 ACTIVITY COMPONENTS.......................................................................................................................................7 DEVELOPMENT HYPOTHESIS ................................................................................................................................8 EVALUATION PURPOSE...........................................................................................................................................9 EVALUATION METHODOLOGY AND STRUCTURE...................................................................................10 KEY EVALUATION QUESTIONS (EQs)..................................................................................................................10 FINDINGS AND CONCLUSIONS ............................................................................................................................11 EQ1: RELEVANCE......................................................................................................................................................11 EQ2: EFFECTIVENESS ...............................................................................................................................................12 EQ3: EFFICIENCY ......................................................................................................................................................23 EQ4: SUSTAINABILITY.............................................................................................................................................26 EQ5: SCALABILITY....................................................................................................................................................28 EQ6: ADDITIONALITY............................................................................................................................................30 EQ7: GENDER.............................................................................................................................................................31 EQ8: YOUTH ..............................................................................................................................................................33 RECOMMENDATIONS.................................................................................................................................................34 EQ1: RELEVANCE......................................................................................................................................................34 EQ2: EFFECTIVENESS ...............................................................................................................................................34 EQ3: EFFICIENCY ......................................................................................................................................................37 EQ4: SUSTAINABILITY.............................................................................................................................................38 EQ5: SCALABILITY....................................................................................................................................................38 EQ6: ADDITIONALITY............................................................................................................................................39 EQ7: GENDER.............................................................................................................................................................39 EQ8: YOUTH ..............................................................................................................................................................40 LESSONS LEARNED ......................................................................................................................................................40 ANNEXES .........................................................................................................................................................................41 ii | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 1. LINQ INCEPTION REPORT .............................................................................................................41 ANNEX 2. EVALUATION DESIGN AND METHODOLOGY ......................................................................42 ANNEX 3. GANTT CHART - EVALUATION WORK PLAN.......................................................................50 ANNEX 4. EVALUATION DESIGN MATRIX ....................................................................................................51 ANNEX 5 FIELDWORK TOOLS...........................................................................................................................52 ANNEX 6. PERSONS INTERVIEWED..................................................................................................................53 ANNEX 7. BENEFICIARY SUMMARY AND SELECTION PROCESSES......................................................56 ANNEX 8. RFA PROCESS AND SELECTION RESULTS ...............................................................................60 ANNEX 9. WAY FORWARD – RECRUITMENT AND SELECTION OPTIONS FOR FUTURE GRANTS & TA............................................................................................................................................................62 ANNEX 10. BENEFICIARY-TRAINEE SURVEY PROTOCOL AND RESULTS ........................................66 ANNEX 11. MEL-INDICATOR PERFORMANCE TRACKING TABLE (IPTT) ANALYSIS ...................70 ANNEX 12. SUPPLEMENTAL QUESTION: TRADE FINANCE....................................................................76 ANNEX 13. POST-BEIRUT BLAST ECONOMIC AND OPERATIONAL OVERVIEW...........................77 ANNEX 14. BIBLIOGRAPHY..................................................................................................................................79 ANNEX 15. DECLARATION ON CONFLICT OF INTEREST .....................................................................80 FIGURES AND TABLES Figure 1: IPTT Progress to LOP Targets....................................................................................................................13 Table 2: Primary Data Collection Approach.............................................................................................................45 Table 3: KIIs with USAID and LINQ Management and Staff .................................................................................53 Table 4: KIIs with Consultants......................................................................................................................................53 Table 5: KIIs with Grantee Applicants and Recipient Organizations ..................................................................53 Table 6: KIIs with Beneficiaries-Trainees and Beneficiaries – COVID-19/Emergency Response.................54 Table 7: Summary of Beneficiaries...............................................................................................................................56 Table 8: Summary of Beneficiaries Selected to Receive Grants under LINQ...................................................57 Table 9: Summary of Beneficiaries Receiving Technical Assistance under LINQ ............................................58 Table 10: Summary of Beneficiaries Receiving COVID-19/Emergency Response Assistance (Intervention 2) under LINQ..................................................................................................................................................................59 Table 11: RFA Evaluation Criteria ...............................................................................................................................60 Table 12: Summary Statistics for RFA-001 and RFA-002.......................................................................................60 Table 13: Summary of Successful RFA Applications................................................................................................61 Table 14: Pre-Selected Applications Not Yet Awarded Grants...........................................................................62 Table 15: Farmer/Trainee Survey Results..................................................................................................................68 Table 16: LINQ Performance Indicators and Corresponding Disaggregation (Y2Q2 and Y2Q3)..............72 Table 17: TA and Training Recipients versus TA and Training Application (Y2Q2 and Y2Q3)...................75 iii | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ABSTRACT The United States Agency for International Development (USAID)’s Lebanon Investment in Quality (LINQ) activity is a $5.9 million activity funded by USAID and implemented by Land O’Lakes Venture 37, which works on increasing the domestic and export sales and competitiveness of value-added agro-food products. LINQ started on September 1, 2018 and will end on November 30, 2021. The project has thus far included a wide range of activities catering to small and medium enterprises (SMEs), agro processors, and cooperatives working in the agro-food industry. This mid-term performance evaluation aimed to assess the performance of LINQ in terms of relevance to a rapidly evolving economic and political context; effectiveness in terms of components and activities in achieving objectives and impact; efficiency in terms of monitoring and evaluation; sustainability going forward; and success in gender mainstreaming. The evaluation team carried out 60 key informant interviews (KIIs), plus an independent analysis of the project’s monitoring, evaluation, and learning (MEL) documents. KIIs were conducted with beneficiaries located throughout Lebanon, with a large proportion concentrated in major agricultural areas (Akkar, Baalbak-Hermel, Bekaa, Mount Lebanon, and North). In summary, LINQ has demonstrated considerable flexibility in responding to the evolving needs of Lebanon’s agri-food sector due to multiple crises and at the same time remaining responsive to USAID development objectives. Overall, LINQ interventions (grant, TA, training) respond to beneficiaries’ needs amid political, economic, financial, and public health crises. However, it is not be possible to determine the effectiveness of this assistance until full production cycles are recorded. To a certain extent, LINQ interventions are considered efficient with some interventions more efficient than others; demand-driven packages of grants and technical assistance are likely to yield better results with some reservations expressed concerning standalone trainings focused on reaching large numbers of beneficiaries. Although all questions of sustainability will be heavily conditioned by the current political, economic, financial, and public health crises affecting Lebanon, the impact of LINQ’s grants and technical assistance (TA) is likely to be sustained. LINQ’s inclusion of women has been limited and subject to the bifurcation of traditional gender roles within the agriculture and agro food processing fields. Recommendations include full and continued USAID support for LINQ, with renewed emphasis on the project’s development and economic growth goals using its current mode of intervention (demand￾driven grants and TA) and the adoption of a more focused approach and a narrower range of activities that supports the grant program in light of the current economic crises. iv | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ACKNOWLEDGEMENTS The authors wish to acknowledge the following individuals for their kind support and contributions to this evaluation: - Charbel Hanna, Carol Brakhya, Elias Haddad, and Steve Morin from USAID/Lebanon. - Rana Taher, Carine Khoury, Harvey Herr, Najwa Andraos, Amer Haddad, and Samar Safar from Social Impact/PMSPL II. - Leah Ghoston and Kimberly Ryan from Social Impact Headquarters. The authors appreciate the willingness of the many individuals who served as respondents within the scope of this evaluation and who generously shared their time and experience at this particularly challenging time for Lebanon’s agri-food sector and agribusinesses. v | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ACRONYMS ADS Automated Directives System AOR Agreement Officer’s Representative APS Annual Program Statement ARE Agriculture and Rural Empowerment BCOV Beneficiary-COVID-19/Emergency Response BG Beneficiary-Grantee BTA Beneficiary-TA Only CDCS Country Development Cooperation Strategy CLA Collaboration, Learning, and Adapting COP Chief of Party COR Contracting Officer’s Representative COVID-19 Coronavirus Disease 2019 DDL Development Data Library DEC Development Experience Clearinghouse DO Development Objective EQ Evaluation Question EQUI® Evaluation Quality Use and Impact F2F Farmer to Farmer FDA Food and Drug Administration FGD Focus Group Discussion FTE Full-Time Equivalent GAP Good Agricultural Practices HACCP Hazard Analysis and Critical Control Points IP Implementing Partner IPM Integrated Pest Management IPTT Indicator Performance Tracking Table IR Intermediate Result KII Key Informant Interview LBP Lebanese Pound LC LINQ Consultant LED Lebanon Enterprise Development LINQ Lebanon Investment in Quality LIVCD Lebanon Industry Value Chain Development LMS LINQ Management and Staff LOP Level of Performance LWA Leader with Associates M&E Monitoring and Evaluation MEL Monitoring, Evaluation, and Learning MSMEs Micro, Small, and Medium-Sized Enterprises PMP Performance Management Plan PMSPL II Performance Management and Support Program for Lebanon II PIRS Performance Indicator Reference Sheet PSD Private Sector Development PYD Positive Youth Development Q Quarter RFA Request for Application RFP Request for Proposal SMEs Small and Medium Enterprises vi | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT SOW Scope of Work TA Technical Assistance TIF Trade and Investment Facilitation USAID United States Agency for International Development USD United States Dollar USG United States Government Y Year 1 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT EXECUTIVE SUMMARY PROJECT BACKGROUND The United States Agency for International Development (USAID)’s Lebanon Investment in Quality (LINQ) activity is a $5.9 million activity funded by USAID and implemented by Land O’Lakes Venture 37, which works on increasing domestic and export sales and the competitiveness of value-added agro-food products in Lebanon. LINQ started on September 1, 2018 and will end on November 30, 2021. LINQ partners with high-potential agribusinesses, processors, and growers of fresh produce to improve productivity, product safety, and quality, while creating new market linkages to promote business and profit growth. LINQ develops Business Action Plans for its partners to identify the resources needed to overcome constraints and increase the income of program participants. These plans help identify the type of assistance required by each business and may include technical assistance, investment grants, and employee training. In addition, LINQ provides training directly to cooperatives and farmers to improve their production practices and quality. EVALUATION PURPOSE AND QUESTIONS This mid-term performance evaluation has assessed the performance of LINQ in terms of relevance to a rapidly evolving economic and political context; effectiveness in terms of components and activities in achieving objectives and impact; efficiency in terms of monitoring and evaluation; sustainability going forward; scalability and additionality of the interventions; success in gender mainstreaming and youth participation; and opportunities for the project’s future performance in light of the current challenges facing Lebanon’s agri-food sector. In addition, the evaluation team included a supplemental, exploratory question to assess the issue of access to trade finance (refer to Annex 12). The evaluation took place between June and August 2020 and was primarily, but not solely, retrospective in nature. To ensure that this evaluation is usable, Social Impact worked with USAID/Lebanon to clarify the purpose, users, and decision-making uses of this study, which ultimately led to an agreed focus on nine evaluation questions (EQs). 1 The key intended users of the evaluation are USAID/Lebanon and LINQ’s implementer, Land O’Lakes Venture 37. In the spirit of the USAID Evaluation Policy, the evaluation sought to provide USAID/Lebanon with concise, actionable recommendations based on findings and analyses to be used by the Mission for future program design or program improvement. The team anticipates that the results from the mid-term evaluation will be used by USAID/Lebanon during its annual Portfolio Review. EVALUATION DESIGN AND METHODS This mid-term performance evaluation applied mixed methods, drawing on secondary and primary data. Primary data included qualitative data collected from appropriate stakeholders and beneficiaries of the LINQ activity. Data collection included key informant interviews (KIIs), conducted subject to approval of appropriate precautions in light of the Coronavirus Disease 2019 (COVID-19) pandemic. 1 The evaluation SOW included EQ9: Way Forward, which effectively encompassed all recommendations for LINQ’s remaining performance period. Accordingly, content for EQ9 has been reallocated under the other EQs. 2 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT FINDINGS EQ1-Relevance: In its design, the LINQ activity is directly relevant and appropriate to Private Sector Development (PSD) Result 1 and Intermediate Result (IR) 1.1 and CDCS IR 2.1. Since March 2020, the requirement of USAID/Lebanon that LINQ respond to COVID-19 and other emergencies has forced LINQ to re-direct a portion of its resources away from the PSD Result 1 and IR 1.1 and Country Development Cooperation Strategy (CDCS) IR 2.1, though this support has helped USAID in responding to the Pillar 4 of the U.S. Supplemental Strategy. LINQ faces a tension between its longer￾term development objectives and the shorter-term humanitarian/livelihoods needs that has been exacerbated by the recent economic crisis in Lebanon. EQ2-Effectiveness: LINQ’s performance to date reflects partial effectiveness in its achievement of some but not all of its 17 performance indicators. Indicator Performance Tracking Table (IPTT) data through Year 2 Quarter 3 (Y2Q3) show that the activity is on track to meet less than half of its performance indicators. A common response from KIIs was that “it is too soon to tell” whether LINQ has reached its intended outcomes as much of its work is ongoing. A bright spot has been the combination of demand-driven RFA and a negotiated Business Action Plan, seen to contribute to the effectiveness of grant assistance to beneficiaries. Multiple beneficiaries highlighted the outstanding need for assistance (including TA) to support marketing and establish market (especially export) linkages. As for training interventions, the evidence for their effectiveness is mixed. The current political, economic, financial, and public health crises facing Lebanon pose a serious challenge to local agribusinesses but offer possible opportunities through export promotion and import-substitution. Assistance under the COVID￾19/Emergency Response was reported to be effective in supporting local food production. EQ3-Efficiency: Data analysis shows a marked difference in the average cost of assistance packages provided per beneficiary under grants ($51,701.11) versus TA and training ($2,784.38). Those beneficiaries-grantees that received assistance outside the demand-driven request for application (RFA) process or that did not receive grants plus technical assistance (TA) (as requested in their RFA application) did not fully agree that interventions were efficient. Early in its implementation, LINQ expanded its reach to beneficiaries to widen the impact, largely through training activities; while LINQ management prefers that training activities be linked to grantees or TA recipients to improve efficiency, this was not always achieved. Stakeholders repeatedly noted the importance of training follow-up to ensure proper application and thus efficiency. The ability of the LINQ monitoring, evaluation, and learning (MEL) team to track changes in sales (domestic and export) has been complicated by the devaluation of the Lebanese Pound (LBP). This may necessitate additional reporting of (changes in) sales volumes, similar to what will be done under COVID-19 aligned indicator “Yield of targeted agricultural commodities among program participants with USG assistance” from the next progress reporting cycle. EQ4-Sustainability: Beneficiaries highlighted the combination of grants and TA in addressing their needs and ensuring that the assistance remains sustainable over the longer term. Beneficiaries expressed confidence in their ability to maintain/operate equipment provided through grants under LINQ, although some anticipate challenges in obtaining imported materials going forward. Conversely, multiple respondents expressed doubts about the sustainability of short-duration trainings that are delivered 3 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT without follow-up support. The emergency assistance offered in response to COVID-19 (Interventions 2 and 3) was acknowledged not to be sustainable over the longer term. EQ5-Scalability: Stakeholders expressed mixed confidence that LINQ could support additional beneficiaries without undermining the competitiveness of existing beneficiaries. Existing beneficiaries’ greatest concerns were over possible support to be directed to their same sub-sector, market, or geography. Time- and resource-intensive TA that requires close collaboration between consultants and businesses would be difficult to replicate at scale without recruiting additional consultants. Conversely, focused workshops or panel presentations reached multiple individuals with limited training resources. Training delivered to individual beneficiaries reached scale in terms of number, but not effectiveness. EQ6-Additionality: Stakeholders spoke positively about the additionality of LINQ interventions, particularly in light of Lebanon’s challenging economic situation. Some respondents indicated that beneficiaries would not have invested in improved production and/or business practices without LINQ’s support; or that the investment may indeed have happened, but at a later stage or at a slower rate. EQ7-Gender: LINQ’s approach to women’s participation in its activities is targeted in two ways (RFA screening for women’s participation within business/cooperative, training of women’s cooperatives), and reflects rather than challenges the bifurcation of women’s and men’s roles in the agri-food sector. Stakeholders reported little to no messaging around women’s participation in activities under LINQ. IPTT data through Y2Q3 show wide gaps in performance across gender: Men are the primary beneficiaries (Indicator 2) and account for the majority of individuals enjoying increased incomes as a result of United States Government (USG) assistance (Indicator 3). EQ8-Youth: A lack of messaging around youth participation in LINQ activities is consistent with the fact that youth were not initially a target population under LINQ and its Cooperative Agreement with USAID did not focus on youth participation. LINQ recently adopted an indicator on youth participation and identified a youth focal point to coordinate youth-related issues under the activity. Youth are reportedly more open to new technologies and management practices conveyed through LINQ training and TA interventions than older adults, with respect both to farm practices as well as food processing techniques. CONCLUSIONS EQ1-Relevance: LINQ has demonstrated considerable flexibility in responding to the evolving needs of Lebanon’s agri-food sector due to economic and public health crises and at the same time remaining responsive to USAID development objectives. EQ2-Effectiveness: It will not be possible to determine the effectiveness of grants, TA, or training assistance under LINQ until full production cycles are recorded, and these are likely to fall outside the performance period, especially for new grants. Beneficiaries’ business performance has been adversely impacted by exogenous factors outside the LINQ activity over the past 12 months. Demand-driven assistance packages have been better received and thus far reported better progress in meeting interim deliverables under the LINQ activity. Beneficiaries (grantees and TA recipients) continue to require assistance to support marketing and establish market linkages with customers in both domestic and export markets. Regarding training, the wide gap in reported effectiveness between self-reported data 4 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT through the farmer/trainee survey and the formal MEL reporting indicators and expert assessments means that it is not possible to determine real effectiveness without further validation. Support packages offered by LINQ under COVID/Emergency Response – Intervention 2 were well received, but not fully responsive to the producers’ needs and excluded key inputs. Gaps in the communication process have introduced some challenges for LINQ activity implementation but opportunities for improvement exist. EQ3-Efficiency: Demand-driven assistance packages of grants and technical assistance are likely to yield better results (better response to the needs of small and medium enterprises (SMEs) and the sector overall) and thus higher efficiency. The LINQ RFA application processes’ focus on grants may unintentionally overshadow the options to provide TA. Training is not considered the most strategic and optimal intervention if not linked to grants, particularly without sufficient demonstration and follow￾up support for trainees through an SME, cooperative, or association. Training that cannot demonstrate impact cannot show an efficient use (cost per impact) of LINQ activity resources. The current MEL reporting may insufficiently capture beneficiary performance after LINQ assistance and may require an atypical adjustment by the MEL manager and beneficiaries for reporting. EQ4-Sustainability: Interruptions to the import of necessary inputs may disrupt the production and/or export potential of several LINQ beneficiaries in the next 12 months. Single, stand-alone training sessions seem unlikely to achieve meaningful, long-lasting change within Lebanon’s agri-food sector. EQ5-Scalability: LINQ has been limited in its capacity to expand grants, TA, and training assistance on a wide scale due to relatively limited budget resources and a modest performance period. LINQ’s capacity to scale assistance through grant support targeting expanded production within the same region or outside may vary according to sub-sector, sub-category, and the SMEs’ local or export market emphasis. LINQ’s demand-driven, focused, and resource-intensive approach to grants and TA that tailors assistance to the exact needs of beneficiaries inherently limited the scalability of this assistance. Conversely, LINQ’s efforts to deliver training to a large number of trainees have not thus far been definitively shown to achieve impact beyond reaching a large number of people, though self-reported post-training survey results are promising. EQ6-Additionality: Available evidence suggests that recipients consider LINQ support to be additional. The additionality of LINQ support has likely been heightened in the past year, due to the inability to access investment finance through the commercial banking sector or to self-finance investments. Evidence of donor shopping and donor dependence undermines evidence of additionality. EQ7-Gender: LINQ does not deliver consistent messaging on the importance of women’s participation to its beneficiaries or stakeholders. LINQ’s approach to gender mainstreaming has been limited to increased women’s participation, yet it reflects at the same time a generally reported bifurcation of women’s and men’s roles within the agri-food sector. EQ8-Youth: LINQ is unlikely to conduct comprehensive performance reporting on youth without an explicit requirement to do so from USAID. However, youth may represent a target sub-population for the introduction of transformative agribusiness models, practices, and technologies in Lebanon. 5 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT RECOMMENDATIONS EQ1-Relevance: In light of Lebanon’s current crisis situation, LINQ interventions should be streamlined to respond to current and evolving needs of agro-processors with emphasis on staying within the PSD IR 1, 1.1 and CDCS IR 2.1. Going forward, the LINQ activity should limit the re-directing of its funds/activities to COVID-19/emergency response. A separate mechanism with dedicated resources should be established outside of LINQ to address emergency response such as COVID-19 or other crises, unless the response directly deals with LINQ’s current beneficiaries (grantees and TA recipients). EQ2-Effectiveness: LINQ should immediately apply for a six-month, no-cost extension with USAID to implement and monitor delivery of grants and TA to beneficiaries. In the remaining performance period, LINQ should focus on resilience of existing beneficiaries and support their continued operation in the market; focus on identifying and fostering market linkages and sales channels that correspond to the comparative advantage and quality-price proposition of its beneficiaries; and assist beneficiaries to identify opportunities to reduce production costs in light of rising costs of imported raw materials. For grants, LINQ should retain a cost-share requirement but discuss with USAID possible adjustment. LINQ should re-review all pending applications received under RFA-001 and RFA-002 to ensure that the requests remain valid, that they reflect the applicants’ needs and business opportunities within the current context, and that the proposals merit investment by LINQ. If the preceding review offers an insufficient pool of applicants to account for remaining grant funds, then LINQ should immediately issue a third RFA or an annual program statement (APS) to solicit additional proposals. Remaining grants to be awarded should follow a demand-driven approach and initial consultation process with LINQ staff, as previously done under the RFA process and Business Action Plan negotiation. For training, USAID should validate its effectiveness with external expert visits/field observations and quality assessments of fresh or processed food products. Under the remaining performance period, follow-up should be provided to address gaps in those trainings already delivered, particularly targeting trainees connected to beneficiaries-grantees. LINQ should provide the full list of all beneficiaries that received assistance under COVID-19/Emergency Response Interventions 2 and 3 to USAID, for possible follow-on assistance through other implementing activities or mechanisms. Given rising prices for raw materials and difficulties in importing inputs, USAID should assess within another activity the viability of supporting an existing local manufacturer to expand the supply and/or quality of food-grade packaging materials for use by local agro-processors; and should position and promote Lebanese processed foods as artisanal products in international markets. EQ3-Efficiency: For future grants, LINQ and beneficiaries should discuss priority needs in consideration of the current Lebanese situation (“nice to have” versus “needs to have”), to be vetted by a technical advisory committee comprised of external technical experts. To have a greater impact through TA, LINQ should make clear that it can award TA under its RFA process – regardless of whether a grant is awarded or not. LINQ should prioritize training for farmers/processors/individuals linked to grantees or potential grantees, rather than independent beneficiaries. LINQ should continue to report against the standard Performance Indicator Reference Sheet (PIRS) indicator: incremental sales in United States Dollars (USD) and average price per unit, disaggregated by domestic and export; if the current indicator does not track actual volumes sold at the time of reporting, LINQ should collect and report this information to account for the rapid depreciation of the LBP. 6 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT EQ4- Sustainability: LINQ should start from now identifying/documenting the potential disruptions to grantees’ supply chains as a result of capital controls and/or other inability to import. It should gather and share information on viable and legal mechanisms to access trade finance to minimize such disruptions. LINQ should use single training sessions strategically and only when needed. EQ5-Scalability: Barring any increase in activity budget, LINQ should continue to focus on delivering demand-driven assistance under grants and TA to qualified beneficiaries to offer the most promising results in terms of effectiveness and efficiency. The focus should not be on scalability of grants or TA under the remaining performance period. USAID should look to LINQ’s experience for learning around scalability under other, newer, larger activities (Agriculture and Rural Empowerment (ARE), Trade and Investment Facilitation (TIF)), especially lessons from firm-level assistance that can then be built upon at the sector or sub-sector level. EQ6-Additionality: LINQ should continue to carefully vet RFA applicants to ensure that beneficiaries are not chronically dependent on donor assistance, and thereby demonstrate and reflect additionality. EQ7-Gender: The USAID Lebanon Gender Assessment should be a guiding document in LINQ’s gender strategy and activities. LINQ should identify opportunities to apply all the recommendations from this document across all stages of activity implementation instead of simply focusing on reinforcing women’s and men’s current roles in the agri-food sector. EQ8-Youth: Under the leadership of its newly identified youth focal point, LINQ should work with USAID to clarify in what ways the project can target youth in its remaining performance period. LINQ and USAID should identify opportunities where youth may represent a target sub-population for the introduction of transformative agribusiness models, practices, and technologies. 7 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT EVALUATION BACKGROUND ACTIVITY BACKGROUND The United States Agency for International Development (USAID)’s Lebanon Investment in Quality (LINQ) activity is a $5.9 million activity funded by USAID and implemented by Land O’Lakes Venture 37, which works on increasing domestic and export sales and the competitiveness of value-added agro-food products. LINQ started on September 1, 2018 and will end on November 30, 2021. LINQ partners with high-potential agribusinesses, processors, and growers of fresh produce to improve productivity, product safety, and quality, while creating new market linkages to promote business and profit growth. LINQ develops Business Action Plans for its partners to identify the resources needed to overcome constraints and increase the income of program participants. These plans help identify the type of assistance required by each business and may include technical assistance, investment grants, and employee training. In addition, LINQ provides training directly to cooperatives and farmers to improve their production practices and quality. LINQ is designed as an activity under USAID/Lebanon’s Private Sector Development (PSD) project. In terms of the Mission’s Country Development Cooperation Strategy (CDCS), the activity is a critical component of USAID/Lebanon’s Development Objective 2 (DO2): Inclusive Economic Growth Enhanced, and therefore both directly and indirectly supports other ongoing economic growth projects in areas like value chain and small business development. Since its inception, the LINQ activity underwent two modifications that increased the obligated amounts and proposed changes in the outputs based on USAID’s request. ACTIVITY OBJECTIVES The Farmer-To-Farmer (F2F) Leader with Associates (LWA) LINQ activity is a 39-month (September 2018 – November 2021), $5.9 million, USAID-funded co-operative agreement implemented by Land O’Lakes International Development. The goal of LINQ is to increase the competitiveness of value-added agribusinesses in Lebanon, thus directly contributing to USAID’s PSD Intermediate Result (IR) 1.1, “increased domestic and export sales,” and PSD Result 1 and Lebanon CDCS IR 2.1, “increased competitiveness of value-added agribusinesses.” LINQ works to achieve its goal by providing direct technical assistance (TA) to agribusinesses and farmers and facilitating technology upgrades by means of investment partnerships and grants, described below. LINQ identifies target agro-food subsectors and partners with agribusinesses therein (primarily growers, processors, and other stakeholders that add value to raw agricultural products) to enhance their growth and contribute to improving their income and increasing their domestic and export sales. LINQ also establishes relationships with different stakeholders that directly or indirectly support the agribusiness sector in order to coordinate and collaborate during implementation. ACTIVITY COMPONENTS LINQ focuses its interventions on three subsectors—fruit trees, food processing, and dairy production—and includes the following three components: Technical assistance: TA may include product innovation development, facilitation of market linkages, improvements to production technology and quality, and analysis of business investment opportunities. 8 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT LINQ also provides TA directly in the form of training for cooperatives and farmers through local or international volunteers, paid consultants, and remote mentoring, including expertise leveraged from Land O’Lakes Inc., a premier U.S. agribusiness. Investment grants: Investment grants partners’ need to become or remain competitive by investing in new equipment and technology upgrades. As part of the grant, the LINQ team and partners co-develop Business Action Plans to guide partners on advancing their businesses. Collaborating, Learning and Adapting (CLA): Under this component LINQ focuses on successes and lessons learned as well as on new practices and technologies and uses this component to transfer knowledge to other stakeholders among subsectors. This is done through technical training, site visits, workshops, and conferences, among other activities. In early 2020 and at the request of USAID/Lebanon, LINQ expanded its assistance portfolio to provide assistance for COVID-19/Emergency Response through three interventions. Intervention 1 targeted existing beneficiaries (grant and TA recipients) to address sanitation and food safety concerns linked to COVID-19. Intervention 2 targeted Lebanese farmers in rural areas through the distribution of inputs (seeds, small tools, and equipment) to support resilience to COVID-19 and other economic disruptions. Intervention 3 targeted vulnerable populations by linking them to food ratio distribution organized by other USAID-funded activities, NGOs, or municipal bodies. DEVELOPMENT HYPOTHESIS The LINQ activity is based on the theory that Lebanese agribusinesses will continue to face constraints in accessing domestic and international markets in the absence of 1) product-development and marketing strategies; 2) sufficient production technology and quality and safety standards; 3) available financing for investments; and 4) strong market linkages. According to LINQ’s monitoring, evaluation, and learning (MEL) plan, the development hypotheses underlying the activity are as such: • If businesses receive tailored TA in quality, safety and productivity, then they will have increased knowledge and apply improved practices in those areas. • If businesses receive co-investment and support to access financing, and have increased knowledge in quality, safety and productivity, then they will invest in improvements to quality and safety equipment and infrastructure. • If farmers receive training in quality, safety and productivity and linkages to buyers, then they will increase their knowledge and apply improved practices in those areas. • If farmers apply improved practices in quality, safety and productivity and linkages to buyers, then they will have more products of higher quality, and higher income. • If businesses have increased knowledge in quality, safety and productivity, better equipment and infrastructure, and access to improved supply of inputs from farmers, then they will apply improved practices in quality, safety and productivity, improve the quality of their goods, and better able to meet buyer specifications. • If businesses receive tailored TA in product development and marketing, linkages to buyers, and are better able to meet buyer specifications, then they will increase their sales, pay higher wages to their staff, and increase their workforce. 9 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT LINQ’s expected achievements over the life of the activity are:2 • 2,800 micro, small, and medium-sized enterprises (MSMEs) including farmers, and other organizations receiving business development services. • 1,850 individuals benefiting from increased income. • 1,050 farmers or microenterprises applying new technologies or management practices. • 20% average increase recorded in domestic and export sales of beneficiaries at firm and farm levels. • $3 million in new private sector investment leveraged. EVALUATION PURPOSE The objective of this mid-term evaluation is to assess the performance of LINQ in terms of relevance to the economic and political context, effectiveness in terms of components and activities in achieving objectives and impact, efficiency in terms of monitoring and evaluation, sustainability going forward, and success in gender mainstreaming. The evaluation has taken place between June 2020 and August 2020. The recent economic developments3 combined with Coronavirus Disease 2019 (COVID-19) challenges pose important questions for LINQ particularly because of its focus on a productive agricultural sector and its requirement for cost sharing, among others. The current situation forced LINQ to shift its activities to address immediate questions associated with food security. Today, LINQ is at a crossroads as the whole model upon which it was designed may no longer be applicable considering the financial crisis and lack of liquidity among small and medium enterprises (SMEs).4 LINQ faces major challenges and obstacles considering a shaky banking and financial sector. This mid-term evaluation has taken into consideration the status of the Lebanese economy in drawing conclusions about project successes and challenges. At the same time, it is primarily retrospective in nature and does not aim to provide speculative answers as to what will happen six months down the line within Lebanon’s social and political landscape. PMSPL II has conducted this evaluation in accordance with Social Impact’s Evaluation Quality Use and Impact (EQUI™) approach, processes, and protocols. EQUI processes are designed to achieve a quality evaluation that incorporates learning and utilization. In addition to assessing program performance in achieving planned objectives, it highlights lessons learned and provides recommendations taking into consideration programming of future activities in Lebanon (consistent with what can be discerned at present concerning Lebanon’s social and political landscape). To support utilization, this evaluation will be followed by a post-evaluation utilization plan. PMSPL II will continue to actively engage USAID/Lebanon, the implementing partners (IPs), and other stakeholders in the evaluation process. In accordance with Automated Directives System (ADS) 205.3.8.2, this evaluation sought to capture the situations and experiences of both males and females and highlight any challenges and/or achievements in 2 Figures are reported in LINQ’s “Annual Progress Report October 1st 2018-September 30th 2019.” 3 Since mid-2019, Lebanon’s economy has faced significant disruption and contraction due to political revolution, currency depreciation, informal capital controls applied by commercial banks, rising unemployment, and physical lockdown and related measures due to COVID-19. 4 Annex 1 presents the Evaluation Inception Report, including critical assumptions underlying LINQ performance. 10 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT filling gender gaps identified in the LINQ gender strategy. It includes the extent to which the LINQ activity meets the gender objectives of USAID/Lebanon’s CDCS as a result of the activities that were implemented. It also looks at youth engagement within its current activities using the gender dimension as young women may differ in their needs and the challenges they encounter compared to young men. The evaluation team consists of Ms. Rachel Bahn, Team Leader and Mr. Bashar Berro, Lebanon Agricultural Expert. Dr. Rana Taher, PMSPL II Senior Monitoring and Evaluation (M&E) Specialist, and Ms. Leah Ghoston, Project Director at Social Impact Headquarters, provided the team with management and logistical support. EVALUATION METHODOLOGY AND STRUCTURE The evaluation combined mixed methods with a strong emphasis on qualitative data. The evaluation team conducted a total of 60 key informant interviews (KIIs) with a combination of stakeholders and end beneficiaries including SMEs and individuals supported by LINQ through technical assistance and training. Annexes contain detailed information on the evaluation methodology including the evaluation design and methodology (Annex 2), the Gantt chart providing the full timeline of the evaluation’s activities (Annex 3), the evaluation design matrix (Annex 4), all fieldwork tools (Annex 5), and the list of persons interviewed (Annex 6). The remainder this evaluation is structured as follows: The next section presents the key evaluation questions guiding this work. Findings and conclusions are then presented by EQ, followed by recommendations. Supporting information including detailed data analysis is located in additional annexes (Annexes 7-15). Findings, conclusions, and recommendations for the supplemental question on trade finance are also located in an annex (Annex 12). KEY EVALUATION QUESTIONS (EQs) EQ1: RELEVANCE How relevant and appropriate are LINQ’s approach and interventions in meeting the objectives and performance indicators for USAID’s Private Sector Development (PSD) Result 1 and Intermediate Result 1.1 and USAID/Lebanon’s Country Development Cooperation Strategy (CDCS) Intermediate Result 2.1)? EQ2: EFFECTIVENESS To what extent are the LINQ approach, interventions, and results effective in meeting the objectives and performance indicators of the LINQ Cooperative Agreement? What were the outcomes? What were the primary factors responsible for their effectiveness? EQ3: EFFICIENCY To what extent were LINQ’s selected interventions and entry points in each subsector strategic and optimal in achieving the intended results? EQ4: SUSTAINABILITY Which activities or interventions were more sustainable than others, and what were the primary synergies that contributed to that sustainability? EQ5: SCALABILITY To what extent were LINQ’s interventions scalable in terms of size and number? EQ6: ADDITIONALITY To what extent do LINQ’s interventions demonstrate additionality? 11 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT EQ7: GENDER To what extent were recommendations generated by the gender analysis (USAID Lebanon Gender Assessment) integrated in LINQ implementation? What are the outcomes generated as a result of LINQ gender mainstreaming? EQ8: YOUTH To what extent were youth involved in the LINQ Cooperative Agreement? What can be done to increase youth mainstreaming in LINQ assistance? EQ9: WAY FORWARD In view of the challenges faced during the first year and a half of activity implementation and of the ongoing economic situation in Lebanon, what can be done to adjust LINQ’s vision and interventions in a way that is most sustainable and efficient? • Due to the nature of EQ9, it effectively encompasses the recommendations for LINQ’s remaining performance period. Accordingly, all content for EQ9 has been reallocated under the other EQs. FINDINGS AND CONCLUSIONS EQ1: RELEVANCE How relevant and appropriate are LINQ’s approach and interventions in meeting the objectives and performance indicators for USAID’s Private Sector Development (PSD) Result 15 and Intermediate Result 1.16 and USAID/Lebanon’s Country Development Cooperation Strategy (CDCS) Intermediate Result 2.17? Activity Goals: LINQ activity documentation including its Year 1 Work Plan states that, “The goal of the project is to increase the competitiveness of value-added agribusinesses in Lebanon, thus directly contributing to USAID’s Private Sector Development (PSD) Intermediate Result (IR) 1.1, increased domestic and export sales, and PSD Result 1 and Lebanon’s Country Development Cooperation Strategy (CDCS) IR 2.1, increased competitiveness of value-added agribusinesses.” Respondents generally indicate that the LINQ activity in its design is directly relevant and appropriate to PSD Result 1 and IR 1.1. and CDCS IR 2.1, because it specifically seeks to increase the competitiveness of agribusinesses including agro processors as well as to increase their domestic and export sales through improvements to production volumes and quality, as well as by supporting market linkages. Emergency Response: Since March 2020, the requirement of USAID/Lebanon that LINQ respond to COVID-19 and other emergencies has forced LINQ to re-direct a portion of its resources away from the PSD Result 1 and IR 1.1 and CDCS IR 2.1 to fund emergency response and provide rapid support to small farmers and households that do not constitute the profile of agribusiness beneficiaries under LINQ. The project helped USAID respond to the Pillar 4 of the US Supplemental Strategy “Prepare for, mitigate, and address possible second-order economic, civilian-security, stabilization, and governance impacts of COVID-19, in part to prevent development backsliding.” 5 PSD Result 1: Private sector competitiveness increased 6 PSD Intermediate Result 1.1: Increased domestic and export sales 7 CDCS Intermediate Result 2.1: Increased domestic and export sales 12 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT • LINQ redirected a total of $80,287.84 of its $5.9 million budget to cover direct costs associated with the COVID-19/Emergency Response.8 Indirect costs including management and staff labor costs are not reflected within this figure. • LINQ financial statements, specifically those for Y2Q1 and Y2Q2, show the first significant revision in LINQ’s budget during the implementation period. The budget for labor increased by more than $275,000, while the budget for grants (direct costs) was reduced by $250,000 (desk review). Neither change clearly corresponds to the additional direct costs associated with the COVID￾19/Emergency response. The respondents highlighted a tension between the longer-term development objectives of LINQ related to the higher-level activity objectives and performance indicators and the shorter-term humanitarian/livelihoods needs, which more closely relate to USAID/Lebanon’s efforts to support inclusive economic growth. This tension has been present from the initial design of the LINQ activity but the recent economic crisis in Lebanon has exacerbated it. CONCLUSIONS LINQ has demonstrated considerable flexibility in responding to the evolving needs of Lebanon’s agri-food sector despite natural, economic, and health crises while remaining responsive to USAID’s development objectives (DOs) and emergency response requirements at all levels (hail storm (apple), economic crises, devaluation of Lebanese pound, COVID-19, CDCS, PSD, Inclusive economic growth). Until March 2020, the LINQ activity in its design directly responded to PSD Result 1 and IR 1.1 and CDCS IR 2.1. COVID-19 emergency responses diverted LINQ’s work to help USAID in responding to the Pillar 4 of the US Supplemental Strategy “Prepare for, mitigate, and address possible second-order economic, civilian-security, stabilization, and governance impacts of COVID-19, in part to prevent development backsliding.” EQ2: EFFECTIVENESS To what extent are the LINQ approach, interventions, and results effective in meeting the objectives and performance indicators of the LINQ Cooperative Agreement? What were the outcomes? What were the primary factors responsible for their effectiveness? For the sake of organization, findings for EQ2 are presented for the objectives and performance indicators, then per intervention (overall, grants, TA and training, COVID-19/Emergency Response). Objectives and Performance Indicators - MEL9 Performance Indicators: LINQ has adopted 17 performance indicators, of which 14 are PSD or CDCS performance indicators. LINQ’s performance through Y2Q3 reflects that LINQ is on track to meet less than half of the final level of performance (LOP) targets for its performance indicators, reflecting gaps in its effectiveness. Figure 1 presents performance against LOP targets, with orange indicating relative achievement as of Y2Q3 against the benchmark of time elapsed. Only four indicators 8 Information provided via email communication with LINQ management, dated August 6, 2020. 9 This section is based on desk review of LINQ internal documents and IPTT data analysis. Refer to Annex 11 for more detail. 13 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT were on track for achievement as of Y2Q3. Notably, LINQ has achieved no results thus far in generating incremental sales at the firm level or at the farm level (Indicators 7 and 8).10 Conversely, LINQ has performed relatively well and ahead of schedule on those four indicators that are outcome indicators, over which it has a greater degree of control (Indicators 2, 9, 10, and 17). LINQ is also on track to meet Indicator 4, the proportion of beneficiaries who are female, which uniquely has an LOP benchmark of 30%. IPTT data do not indicate any results (“N/A”) for Indicator 6 (macroeconomic/financial situation) or Indicator 16 (subsidized loans accessed through banks/financial institutions) (desk review – refer to Annex 11 for detail). Figure 1: IPTT Progress to LOP Targets Investment Leveraged: LINQ has been more effective in leveraging non-capital investment by the private sector than capital investment (Indicator 1). While LINQ’s performance for leveraging non￾capital investment is “ahead of schedule,” the capital investment leveraged is lagging behind. In absolute terms, LINQ had leveraged more than $1 million in non-capital investment as of Y2Q3 (against a target of $1.5 million), compared to more than $300,000 in capital investment (against a target of $1.5 million). This finding is further supported by the lack of results against Indicators 14 and 15, which indicate that no beneficiaries have been able to access alternative financing sources (commercial loan, private loan, or private equity) (desk review). Activity Expenditures: LINQ financial statements indicate that approximately 50% of the $5.9 million in total activity funds (direct amount) have been expended through Y2Q2, as well as 48% of funds for labor and 50% of funds for grant investment. This figure is slightly below the proportion of the activity’s implementation period that had elapsed (22 of 39 months, or 56%) (desk review). Please refer to Annex 11 for the detailed analysis of LINQ objectives and performance analysis, beyond the key points highlighted above. 10 Indicator 7 is reported semi-annually, Indicator 8 is reported annually. 14 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Outcomes – Overall Demand-Driven Approach: Respondents noted that the combination of a demand-driven proposal and an initial consultation process between LINQ staff and beneficiaries (SMEs and cooperatives) that consisted of proposal review and technical meetings to determine the Business Action Plan resulted in a common understanding of the most relevant interventions to respond to the beneficiary’s needs. However, beneficiaries that were approached by LINQ outside the RFA process – notably those selected in response to the June 2019 hailstorm in Akkar and North Lebanon that damaged fruit orchards – report that they had a more limited role in determining the content of grant support. • Multiple beneficiaries receiving assistance in the form of grants and/or TA (not training) were selected outside the RFA process as designed in order to quickly respond to business opportunities and/or natural disasters (see also Annex 7). “Too Soon to Tell”: A common response was “It is too soon to tell” whether LINQ has reached its intended outcomes, because its work is ongoing: Many grants have not yet finished (equipment not yet installed, expansion not yet complete), the production process is still underway, or the TA or training has not yet been applied to the fullest. Interim variables may have been achieved (e.g., establishing a website, designing a new label), particularly those identified through the demand-driven Business Action Plan, but objectives will not be reached until true application leading to additional sales, new markets, new product, etc. This is reflected in the extension of many grants until end-2020. Continued Need for Market Linkages: Multiple beneficiaries expressed the belief that assistance (grant, TA, or training) to increase competitiveness and production capacity is of limited effectiveness without assistance to support marketing and the establishment of market linkages. LINQ Support amid a Challenging Business Context: While some beneficiaries’ performance targets were not yet achieved, this was most often attributed to the very challenging economic situation facing Lebanon or the effects of COVID￾19. While few beneficiaries spoke to delays linked specifically to internal business issues or to LINQ implementation, other beneficiaries spoke to the role of LINQ in keeping their business profitable or operational during the economic crisis. Business Opportunities and Response to Economic Crisis: Several respondents indicated that the economic situation in Lebanon would worsen in the coming months. Respondents noted that the current economic and financial crises facing Lebanon offer possible opportunities for local agribusinesses to gain competitiveness and market advantages in two flavors: export promotion and import-substitution. These opportunities exist, but they may not be easy to seize. Businesses may require a detailed review and revision of product costing and pricing structures to gain export advantage, for example. Multiple beneficiaries reported that they “Especially with the coronavirus, with the lockdown . . . it was really difficult to call a farmer and tell him, we want to visit you, we want to visit your farm. You can understand that they don’t want now. . . To call now the farmers for a training indoors? They will not come” (Beneficiary-Grantee, Fresh Fruit Sector). “Without the grant, we could not do anything that we are currently doing. We would never be able to continue – we would have shut down” (Beneficiary￾Grantee, Food Processing Sector). “Today we have an opportunity – unfortunately coming out of a catastrophe. But opportunity doesn’t mean it’s easy, it requires effort and linkages and marketing and agricultural producers” (LINQ Consultant, Food Processing Sector). 15 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT saw export as the only viable way to conduct their business at this time, as it is the sole source of “fresh money.” Beneficiaries generally expressed a clear preference for export as opposed to import substitution. Others noted that while export was not the only viable market channel, the inflow of “fresh money” would help them to manage their cash flow. Already some beneficiaries reported that they have responded to the economic crisis by streamlining their operations, focusing on their core business, and reducing excess capacity, others have focused their efforts on reaching export channels. Input/Supply Disruptions: Several beneficiaries noted that maintaining their supply of raw materials had become more difficult in recent months, particularly for imported packaging materials including glass jars, plastic bags, cardboard packaging, and labels, due to reduced access to bank deposits and specifically hard currency. A stakeholder clarified that there were two local manufacturers of glass jars in Lebanon, but supply is otherwise imported from Egypt or Syria. While there was no shortage of supply on the market at the time of the KII, prices were going up, in line with general inflation in the country. • Beneficiaries in the dairy sector noted a lack of sufficient supply of goat milk on the market, which drives up their production costs or limits their operations. This is the result of the purchase of goat herds in recent months, in reaction to fears over food insecurity. COVID-19 Challenges: Beneficiaries that depend on sales through market channels including farmers markets, exhibitions, and agro-tourism reported a decline in sales as a result of COVID-19 and the associated lockdown that banned or discouraged public gatherings. General Support but Few Specific Proposals to Advance Agri-Food Sector: The respondents uniformly supported the continuation of support to Lebanon’s agribusinesses by USAID. There was no consensus on how to further develop Lebanon’s agri-food sector. While a small number of respondents argued that material assistance to the sector should be widened to reach farmers or primary producers more explicitly, the consultants had a tendency to recommend additional support in their respective area of expertise, such as focusing on local raw and input materials, support marketing and packaging, improve cooperative performance through stakeholder FGD. Prioritizing linkages to export markets and access to “fresh dollars” as opposed to the provision of equipment and machinery was reiterated by beneficiaries. Opportunities for Value-Added Production and Promotion: Lebanon’s agro-processing sector faces difficulties in competing with other countries due to the high costs of inputs like energy and the limited scale of production, given the country’s relatively small size of agricultural lands and agricultural production capacity. However, opportunities for value-added production and promotion exist. There is an opportunity for small dairy producers to process milk into cheeses, earn a higher return, and have a positive net impact on rural incomes. However, producers have not taken this step because they are not aware of this market opportunity and they lack knowledge of cheese making. One beneficiary has proposed to support these dairy producers to train them in cheese-making. In addition, the beneficiaries noted an opportunity for Lebanese processed foods (including dairy products like artisanal cheeses) to be marketed as artisanal products, in part because the scale of production may not favor a cost￾competitive marketing strategy. Agro-processing is a chain of consistent production, high capacity automated processing, developed logistics, and advanced marketing. Desk review and evaluation team expertise confirm that, while many developed countries have competitive value chains, Lebanon faces shortages in production to meet 16 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT processors’ needs, lacks the capacity to install advanced processing lines, and suffers from insufficient logistics and marketing. As for export, Lebanese exports typically route through distributors who sell those goods to the catering/restaurant sector or retailers. In most export destinations, there are a large number of manufacturers and strong, local competition that are better able to secure promotions and sales/price positioning. Lebanese exports are small in volume and not sustainable or competitive in export markets unless they pursue the niche sales position. This is complicated by Lebanon’s political, economic, and recently security issues that have impacted trade through closure of land borders with Syria, limited capacity at Beirut airport, and the recent destruction of the Port of Beirut (desk review). Outcomes - Grants Satisfactory Procurement Process despite Challenges: Under grants, the process for identifying appropriate equipment needs, the correct manufacturer/supplier, and securing delivery takes time even in normal circumstances unaffected by COVID-19 or informal capital controls. Additional time is then needed to train staff and implement and test the new production processes/lines. Despite challenges related to the procurement of equipment and machinery (domestic or imported) in the past 18 months, beneficiaries made no/very few complaints about delays associated with the procurement and delivery under the grants. Demand-Driven Grants Achieving Outcomes: Beneficiaries have generally reported that grant support was effective in responding to their business’s/cooperative’s needs, specifically in that the grant achieved its immediate purpose (e.g., procurement and installation of needed equipment or machines), particularly those needs identified through the demand-driven Business Action Plan. In some cases, the grant support has thereby achieved the intended objectives of expanding production volumes, increasing production quality, increasing sales, or introducing new products to the market. • A few beneficiaries reported increased export sales as a result of the assistance provided by LINQ or subsequently. • A small number of beneficiaries said the type of grant assistance provided by LINQ did not always and exactly correspond to their self-perceived needs (one beneficiary received pruning shears it did not request; another requested a vehicle, wood shredder, and weeder but these were not provided). In another case, the beneficiary received equipment that corresponded to his needs but found its corresponding co-investment to be too expensive and not profitable. “The delay in delivering the equipment was force majeure, it was under no one’s control but was due to the closure of companies [COVID-19 lockdown] imposed by the government” (Beneficiary-Grantee, Food Processing Sector). “LINQ assistance has lifted the brand. We introduced a new brand for [company], now we are able to export it. Before we were not exporting. We improved the brand and also the quality through a new machine that can test olive oil quality, so we are more selective on the quality. Now we export to Dubai and Europe. Our production volume has doubled” (Beneficiary-Grantee, Food Processing Sector). “Before I had sent samples to the Gulf but did not yet have a sales contract at the time I started working with LINQ. Now I have an export purchase order to Oman” (Beneficiary-Grantee, Food Processing Sector). 17 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Cost-Share Challenges: USG guidance and regulations strongly recommend/require a cost-share as a way to ensure cost-effectiveness. Existing cost-share practices within LINQ appear to aim for 1:1 cost￾share (document review). However, in light of the difficult and worsening economic situation in Lebanon, multiple stakeholders (including beneficiaries) suggested that the cost-share requirement could be reduced from 50-50 to a ratio more favorable to beneficiaries. Future Grants – Budget: LINQ has extended 19 grants thus far towards its target of 30 grants total, with approximately $400,000 remaining in its budget to support grants. In the remaining performance period, the LINQ team may restrict selection of beneficiaries for grants and technical assistance to those applications submitted to the first two RFA rounds as a means to reduce the time needed for review and implementation. The LINQ team has expressed confidence that these applications are competitive and not “second-choice.” However, since RFA-001 closed in February 2019, and RFA-002 closed in August 2019, Lebanon’s agro-processing sector has faced considerable pressures as a result of political, economic, and public health challenges. • The eight (8) applications received under RFA-002 that passed the pre-selection screening but that have not been provided any grant support to date, requested support totaling $529,665 (data analysis). This figure exceeds LINQ’s remaining budget for grants. • On average, grantees have received LINQ funding contributions that are approximately equal to the amount requested within the RFA application (lower by $288.33, across the 12 applications received and funded under RFA-001 and RFA-002 thus far) (data analysis). Future Grants – Selection: The defined, one-month application window for RFA-001 and RFA-002 was cited as problematic by two beneficiaries. One successful beneficiary noted that the award of grant support under the RFA did not align well with seasonal production schedules. At least one beneficiary was disqualified from receiving grant assistance because the application was filed after the closure of the RFA window. Management-level stakeholders noted that an APS mechanism would allow for a longer application window, allowing potential beneficiaries more flexibility in submitting their applications to LINQ. However, LINQ management noted that an RFA was preferred over an APS to facilitate management, as it allows LINQ staff to bundle and review applications along a consolidated timeframe. No-Cost Extension: At least one respondent spoke clearly and forcefully of the need for a no-cost extension for the LINQ at the activity level to complete the delivery of assistance to grantees, in light of multiple disruptions in the past year. LINQ has offered extensions of approximately 6 months until end￾2020 to multiple grantees, due to disruptions linked to COVID-19 and Lebanon’s economic challenges within the past 12 months. 18 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Outcomes - TA & Training Effectiveness of TA: Beneficiaries’ perceptions of the effectiveness of technical assistance has varied. Whereas some beneficiaries spoke very highly of the content and quality of technical assistance provided by consultants, and confidence that it had directly contributed to their business performance; others expressed disappointment and doubt that it had made any meaningful contribution to their business. Other beneficiaries expressed mixed perceptions: One agribusiness that benefitted from more than one type of TA delivered by different consultants/trainers/volunteers found them to be of varying usefulness. In another example, two businesses that received similar TA from the same consultant offered divergent assessments of its effectiveness. • Beneficiaries within the dairy sector expressed general satisfaction with the TA provided thus far, covering topics including cheese production and quality, food safety, and sales and marketing. Effectiveness of Training: LINQ has delivered extensive training, particularly to the fresh fruits sector and to food processors,11 but evidence of its effectiveness is mixed. Farmer/trainee survey results show that the most prevalent type of training delivered was on Good Agricultural Practices (GAP)- General (81%), followed by integrated pest management (IPM) (59%) and food safety and food processing (14%) (data analysis – see Annex 10 for further details). While beneficiaries including farmer clusters or cooperatives expressed that training offered thus far is traditional or classical in content, cooperative members benefitting from food processing trainings expressed that they had learned something valuable in terms of new techniques or new procedures. • Farmer/trainee survey results report a significant majority (more than 70%) of trainees have applied new technologies or management practices as a result of the training provided through LINQ assistance. Among those trainees who had not implemented the techniques or practices covered in the training and who offered a substantive explanation as to why, the largest proportion attributed this to the timing of the training (data analysis – see Annex 10 for further details). Conversely, one consultant estimated that only 20-25% of farmers trained were applying the training as delivered, a figure well below the self-reported rate of trainees. • Several beneficiaries noted that techniques and management practices for which training was delivered may not be implemented by all farmers, due to a lack of relevant equipment, specifically on tree pruning and plowing. For example, a large cooperative with hundreds of members received pruning shears that are useful, but insufficient to serve all members who were trained in improved pruning techniques given the limited 60-90 day pruning period. 11 There are no survey results from the dairy sector, and this appears to be because no such trainees were surveyed in Y1 (see Annex 10). LINQ management subsequently confirmed that training specific to the dairy sector had reached only eight (8) individuals (email communication dated August 6, 2020). “The only assistance we received from LINQ was not so satisfying comments on the labels, and that was it. . . There was a consultant that you hired [who] helped give us comments on the labels that we had, but those comments were in some points contradictory to the lady that gave us the lecture or workshop at the NDU. It was very slow, it was very inefficient and limited” (Beneficiary, Food Processing). 19 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT • At least one respondent noted that she had attended multiple training sessions on the training topic or similar topics, making it difficult to isolate the impact of any single training session. Conversely, the same beneficiary had received training on grape pruning techniques and applied it not to grapes but to her almond trees, noting a positive impact on fruit size this season. Economic Situation Limits TA/Training Effectiveness: Two beneficiaries noted that some aspects of the technical assistance or training would have been (more) effective, were the business situation “normal” and not affected by disruptions including the economic crisis or COVID-19. For one of these, in light of disruptions to business practices due to COVID-19, the technical assistance can have little practical effect on his business operations. Follow-Up Needed: Respondents (both beneficiaries and consultants) repeatedly spoke about the importance of follow-up to both technical assistance and training. Consultants in several cases provided follow-up even after their work formally concluded (at no cost), in response to beneficiaries’ requests. This insistence on follow-up appeared to be particularly relevant in the case of trainings delivered over a short period of time to a large number of trainees that allowed for minimal time for application in the presence of the trainer/consultant. Opportunity for TA Learning: Several consultants indicated that their contact with beneficiaries had ended after the limited consultation period. While this is standard practice in consultancy work, they therefore had no information on whether or not the guidance provided had been applied, the intervention had been successful, or new challenges had been encountered. This finding is in stark contrast with the routine contact maintained by the LINQ team and the beneficiaries-grantees (once a week or once biweekly during delivery of assistance). Outcomes - COVID-19/Emergency Response Intervention 1: Beneficiaries who received in-kind assistance (masks, gloves, aprons, sanitizers) expressed their satisfaction with and appreciation for the materials they obtained. For example, a beneficiary who received a larger piece of equipment for sanitizing the production facility noted that this machine had reduced the time for disinfection and offered a higher quality of disinfection. Beneficiaries similarly expressed general appreciation for the TA they received on issues of hygiene and food safety linked to COVID-19, specifically through an online workshop/consultation meeting with a food safety and hygiene specialist. However, some beneficiaries expressed that the information shared was basic or remedial and therefore made no meaningful impact on their production practices. Intervention 2: Beneficiaries receiving seeds and seedlings expressed that this assistance was effective in supporting local food production, which has been harvested and dedicated to household food consumption. Potato seeds were preferred to vegetable seedlings. Beneficiaries reported that seedlings/seeds were more productive in the presence of fertilizers or compost and noted the need for fertilizers in addition to planting material. One beneficiary proposed training in compositing could be “[Due to the branding and social media marketing workshop] we had more contacts from new clients, so more sales. . . If it was a normal situation in our country, it might have been 25-30% more sales, but because the situation is bad, we had an increase of maybe 10% through social media platforms” (Beneficiary-Trainee, Food Processing Sector). “If it was normal, if I was participating in an exhibition, [the brochure] would be very, very helpful. But now the situation is like . . you know about the situation” (Beneficiary-TA Only, Food Processing Sector). 20 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT delivered to meet the need for fertilizers locally. Beneficiaries expressed that they needed sanitizers but did not receive any. Extension of Emergency Response: Several members of the LINQ management and staff noted that LINQ may plan to extend some components of the emergency assistance that it provided in response to COVID-19, understood to refer to distribution of planting materials under Intervention 2. Outcomes - Communications Communication Challenges: Beneficiaries revealed three instances where communication with the LINQ team is sub-optimal, undermining its ability to achieve its objectives. First, beneficiaries more readily identified the technical team members than the LINQ project. Second, beneficiaries also report that communications tend to be routed through a small number of technical team members, which imposes a high burden on those members to respond promptly to every inquiry. A small number of beneficiaries noted extensive delays and/or a lack of communication related to the RFA review process and notification of final decisions. In at least one case, communications between the beneficiary and the LINQ team may have been interrupted by staff turnover, contributing to unclear status: For example, an agro-processor was identified as a beneficiary for LINQ assistance, however, according to the beneficiary he has been disqualified from receiving any assistance whereas LINQ documentation still lists this beneficiary. Third, many beneficiaries conflated USAID, LINQ, and (to a lesser extent) Farmer-to￾Farmer and referred to them interchangeably. This may have been exacerbated by an atypical data collection protocol for this mid-term performance evaluation, as the evaluation team rather than the LINQ implementing team was the first point of contact for arranging KIIs with many (most) stakeholders. • Individual or smaller beneficiaries (cooperative members, individual trainees) appear to have no idea about what USAID is, what it does, and how it subcontracts companies to implement activities like LINQ; or about the hierarchy of such activities or that their staff are temporarily working for the activity (may have worked on previous or future but separate activities). Beneficiaries who also received assistance from LIVCD appeared to assume that LINQ is LIVCD 2 because of the long time that key LINQ staff had spent with them under the earlier activity. CONCLUSIONS Overall LINQ interventions have achieved some outcomes, but do not yet demonstrate effectiveness. Qualitative evidence suggests that certain key performance indicators that are only reported annually or semi-annually (Indicators 7 and 8) may report some effectiveness with updated reporting as of Y2Q4. Until full production cycles are recorded, it will not be possible to definitively Distribution of seedlings in Mristi, Shouf in April 2020 under the COVID-19/Emergency Response. (Source: LINQ) 21 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT determine the effectiveness of grants, TA, or training assistance under LINQ (competitiveness, sales, employment, and innovative products). Full production cycles are likely to fall outside the performance period, especially for any forthcoming grants. Nevertheless, LINQ seems unlikely to achieve targets to leverage private sector investment or facilitate commercial loans, given the current status of Lebanon’s financial sector: While these targets may have been ambitious at the outset of LINQ’s performance period, they now appear impossible. Lebanon’s crises – including political revolution, economic contraction, financial sector collapse, and COVID-19 and associated lockdown – have had a clear and largely detrimental effect on beneficiaries’ business performance over the past 12 months. COVID￾19 has had a particularly negative impact on local sales of cooperatives and small enterprises because many sales channels were blocked and closed. Some beneficiaries have responded to Lebanon’s economic crisis by streamlining their operations, while others did so by focusing on export. Despite the crises, LINQ’s staff have been able to mitigate the challenges associated with the sourcing of machinery and equipment for grant recipients, delivering them generally on time and in line with the beneficiaries’ expectations. LINQ assistance could be adapted to the current business context. The LINQ team did not clearly identify any concrete plans to respond to the crisis (current strength or more intense) within its core grants, TA, or training activities in the near future in the face of expectations that the economic crisis would worsen in coming months. However, LINQ assistance can be provided along different timelines, reflecting their respective time requirements: For example, LINQ can deliver some assistance to beneficiaries via TA—for example, such interventions as social media marketing training—more quickly than the equipment installation process. • Supporting market linkages was and remains a priority need for beneficiaries. Beneficiaries (grantees and TA recipients) continue to require assistance to support marketing and establish market linkages with customers in both domestic and export markets. • Input supply has become problematic for some beneficiaries in recent months due to reduced access to hard currency to finance imports and outright supply shortages. Beneficiaries may be looking to identify alternative suppliers of imported inputs, that can deliver at lower cost and/or that can accept payment in local currency. Dairy producers who rely on goat milk production have faced a shortage of supply in recent months due to food security concerns, which has led to operations below full capacity. • Cooperatives that highly depend on seasonal sales channels may face a particularly challenging period over the next six to 12 months as a result of the current public health and economic challenges. Cooperatives rooted in agricultural and agro-processing activities typically face seasonal lulls during winter, with reduced activity and sales. Cooperatives depend on events like local exhibitions, ceremonies, and religious feasts to display their goods, make sales, and earn returns in a very short period of time. These events are typically not held during winter. Cooperatives that had expected to make their sales and earn income in summer 2020 instead faced declining per capita income due to limited opportunities to display and sell as well as the LBP depreciation that reduced the real value of their sales. Accordingly, this seasonal lull is 22 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT expected to be worse in the coming year as a result of COVID-19, Lebanon’s economic contraction, and the reduction in tourism, increasing the needs of such cooperatives. Beyond the crisis context, Lebanese agri-business offers value-added rather than cost￾competitiveness. The country’s comparative advantage may be on quality (quality level or quality-to￾price ratio) or niche, value-added products (e.g., apple vinegar and apple cider within the apple value chain, artisanal cheeses within the dairy value chain), rather than on absolute cost. This is attributable to the high costs of key inputs and the limited potential for scale facing Lebanon’s food processing and manufacturing sector. A specific, longer-term challenge for the dairy sector is that small dairy processors may lack the knowhow to process their milk and make it an added value product. Lebanese exports are small in volume and not sustainable or competitive in export markets unless they pursue the niche sales position. Grants Demand-driven grant assistance packages have been better received and have thus far reported better progress in meeting interim deliverables under the LINQ activity. Total grant support may fall short of the 30-grant target, and selection of future grants requires careful re-review of those applications already received and reconsideration of the solicitation mechanism for new applications. LINQ is unlikely to meet its total grant award target (30 grants) by making awards only to those applications pre-selected under RFA-002: These are only eight in number and have requested a total budget that exceeds LINQ’s remaining grant funding. However, applications already received under RFA-001 and RFA-002 may no longer accurately reflect those agro-processors' current situation and most pressing business needs in light of the evolving situation. A switch from an RFA to an APS for the selection of final beneficiaries to receive grant support could avoid unfortunate outcomes in which otherwise qualified beneficiaries are disqualified because of a missed application timeline. An APS may be better suited than an RFA to drive potential applications to follow a standard application process, even if they require more time or assistance to complete the standard application. While an APS approach may be preferable to another RFA for the flexibility in reaching a wider number of beneficiaries on a time cycle that better fits their needs (seasonality), it imposes a practical trade-off in terms of management: It would require LINQ staff to manage communications to beneficiaries without the benefit of a single timeline. LINQ appears to have sufficient budget remaining to support a six-month, no-cost extension beyond the current performance period scheduled to end on November 30, 2021. This extension reflects approximately the same amount of time that was provided to some beneficiaries￾grantees to account for COVID-19 and difficult economic circumstances. Under the grants assistance, a cost-share requirement could be maintained to meet minimum USG requirements and best practices for ensuring cost-effectiveness and beneficiary buy-in, but made less stringent to reflect local economic conditions. 23 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT TA & Training LINQ is overlooking an opportunity to learn from TA performance. Consultants who are not informed of the ultimate contribution of their TA to the beneficiary’s business may be losing an opportunity to improve or refine the services they offer to Lebanese agro-processors more generally. Further assessment is needed to determine effectiveness of training interventions. There is a wide gap between self-reported data through the farmer/trainee survey and the formal MEL reporting indicators on the application of training delivered with USG assistance. Training courses appear to have a good record of converting to improved technologies and management practices, based on available, self-reported evidence. Expert assessments are more modest. Without a validation, it is not possible to determine real effectiveness. Farmer/trainee survey responses suggest that a significant challenge to implementation of practices covered in training sessions is the timing of the session. Moreover, training results are not available for all sectors; it is difficult to assess the effectiveness of TA/training delivered to the dairy sector as they have been excluded from the Y1 training surveys. COVID-19/Emergency Response Support packages offered by LINQ under COVID/Emergency Response – Intervention 1 and Intervention 2 were well received, but Intervention 2 was not fully responsive to producers’ needs and excluded key inputs. Communication The current communication protocols between LINQ staff and RFA applicants have in some cases led to sub-optimal outcomes. Gaps in the communication process have introduced some challenges for LINQ activity implementation, but opportunities for improvement exist. Moreover, the exceptions to the RFA process may complicate the communications between LINQ and its (potential) beneficiaries, and potentially undermine the public perception of values including transparency and fairness that are espoused by USAID and the USG. EQ3: EFFICIENCY To what extent were LINQ’s selected interventions and entry points in each subsector strategic and optimal in achieving the intended results? For the sake of organization, findings for EQ3 are presented per intervention (overall, grants and TA, training, overall-MEL). Overall Mixed Views on Efficiency: Beneficiaries expressed mixed feedback on whether LINQ’s interventions – grants, TA, and/or training – were strategic and optimal in achieving the intended results. Whereas some recipients found them to be “strategic and optimal” in response to the standard wording of the interview question, not all beneficiaries expressed real agreement when they elaborated on this question. 24 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Varied Assistance Costs: Data analysis shows a marked difference in the average cost of assistance packages provided per beneficiary – grants average $51,701.11 while TA and training average $2,784.38 (data analysis). Grants & TA Smaller Grants in Future: LINQ’s staff indicated that grants issued within the remaining performance period might be of a smaller size, as a means to maximize the number of grants. It is unclear whether this smaller grant size reflects a pre-screening of the grant requests submitted under the previous RFA rounds. Preference for Larger Grants: Stakeholders noted that a large number of small grants had less impact than a smaller number of grants targeting key actors within the supply chain. Key actors may have an advantage in two ways: First, they may be able to deliver a centralized point of contact when dealing with export markets (e.g., agents or aggregators). Second, key actors may offer a wider range of products to potential purchasers that can give them a stronger position as a key supplier or that can reduce transaction costs, as is well-known among retail procurement agents (ET technical expertise). RFA Selection and Business Action Plan Support Efficiency: Under the RFA process, the Business Action Plan is developed as a way to identify the best response to the beneficiaries’ business needs and deliver assistance that strategically responds to those needs. Beneficiaries who made proposals to LINQ outlining their needs more clearly expressed more satisfaction with the assistance received. Conversely, beneficiaries who received assistance that was not demand-driven or beneficiaries who did not receive grants plus TA as requested did not fully agree that interventions were efficient. Those beneficiaries that received assistance that was reported to be less demand-driven were clustered in the fresh fruit sector. This was a result of LINQ selecting beneficiaries in the fresh fruit sector following the hailstorms of June 2019, which followed a non-RFA recruitment process and does not appear to be an issue relevant to the sector in and of itself. Some grantees selected outside the RFA process appear to lack the necessary business skills to successfully implement the grant within their SME/cooperative/association, for example beneficiaries who have little or no professional background in the agri-food sector, and TA to address these gaps was not included within their assistance package. Efficiency of Grants: Several beneficiaries received grant assistance to support facility upgrading that does not necessarily correspond to either the production environment (mismatch between facility installed and the seasonal weather conditions) or the comparative advantage of the business (mismatch between short- and long-shelf-life products). Similarly, several beneficiaries received grants for equipment that does not deliver enough capacity to adequately respond to seasonal demand for their respective services (e.g., pruning, spraying). Grant vs. TA Efficiency: A few respondents noted directly or strongly implied that grants were more efficient than TA, based on their experience as well-established business owners who can accurately identify most pressing business needs requiring investment. At least one beneficiary preferred to receive support only in the form of TA as proposed by LINQ, rather than nothing, aiming to later extend the support to reach their “real” needs under a grant. 25 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Some beneficiaries who received grants did not receive TA or received TA that did not correspond to their most pressing business needs. Multiple beneficiaries who had applied for full packages of grant and TA but received only TA expressed a belief that the assistance delivered was not optimal in achieving their business needs. Similarly, some beneficiaries who received training only expressed that training alone was not the optimal intervention in their case, as they needed expert TA or material support (e.g., jars) to improve their production and sales. Training Efficiency of Training: Early in the LINQ implementation period, USAID specifically requested that LINQ expand the number of individual beneficiaries under the activity to widen the impact across a wider population, and this was largely achieved through training activities. LINQ’s management generally prefers that training activities be linked to grants or TA recipients, which reflects the activity design and is “embedded within the selection criteria.” Indeed, training is seen as more efficient when delivered to trainees connected to an SME, cooperative, or association that is committed to implement the content of the training. However, not all training could be connected to grant and TA recipients. Exceptionally, one respondent expressed concern that the inclusion of activities to provide training to a wide number of individuals would have limited impact and therefore yield low efficiency (cost per impact) as an intervention strategy. Follow-Up for Greater Efficiency: Multiple stakeholders highlighted the importance of training follow-up to ensure that training messages are correctly understood and properly applied, to ensure their impact and thereby their efficiency. Training sessions for farmers, food processors, and other trainees was usually delivered in-person in a single, condensed session of several hours over one or two days. Youth vs. Adult Trainees: Many trainees are older adults (not youth) who are established in their production or processing techniques and therefore less willing to adopt new production or processing techniques, particularly within a single training session. This response was not specific to any given sector, but rather a general comment about willingness to learn and change. Overall MEL Limited Role for Volunteers: LINQ IPTT data through Y2Q3 show that LINQ has not made extensive use of volunteer experts to deliver training (Indicator 10), despite the fact that the use of volunteers could represent an opportunity to deliver training at lower cost and thereby contribute to greater efficiency of USG assistance (desk review). Sales Reporting Challenges: Current MEL reporting targets include changes in sales volumes, measured in currency terms (USD or LBP); changes in production volumes; and changes in employment Q: In your opinion, was technical assistance the best means to address the needs of your business? A: “Not really, no. I told LINQ a long time ago, when they approached me, that supporting an SME like [my company] or others would have to be not in a specific, small TA. It has to be holistic. You did a great website, but where is the expert who can link me to exporters? Where is the van to transport more goods to Beirut? It has to be holistic support, not only on one thing” (Beneficiary-TA Only, Food Processing Sector). “A good training is the one that matches the theoretical and the practical. Both together is a good result” (Beneficiary-Trainee, Fresh Fruit Sector). 26 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT (document review - MEL Plan). Because of the economic crisis, and specifically the devaluation of the LBP on the open and black markets (from the official rate of 1,500 LBP/USD to 8,000 LBP/USD as of end-July 2020), changes in sales volumes measured in currency terms may distort real changes in sales volumes in terms of units sold. The ability of LINQ’s MEL team to track changes in sales (domestic and export) has been complicated by this devaluation because reporting of (changes in) sales values obscures (changes in) sales volumes, especially as the exchange rate at which the value is reported is not currently collected from the business. However, volume-based MEL reporting will be done under LINQ as of the quarterly report for Y2Q4 under COVID-19 aligned indicator “Yield of targeted agricultural commodities among program participants with USG assistance.” Post-Training Survey Results: LINQ has conducted ex-post surveys of beneficiaries who participated in training sessions, as a means to document results and track the progress of its activities, specifically against the indicator “Number or farmers or microenterprises who have applied new technologies or management practices as a result of USG assistance.” There are opportunities for improvement in this data collection, as well as deeper data analysis (refer to Annex 10). CONCLUSIONS There is no consensus on which intervention package (grants, TA and/or training) is considered the most strategic and optimal in achieving the intended results. However, the combination of grants and TA—if demand-driven (rather than supply-driven), requested by experienced SMEs, agreed under the Business Action Plan, and negotiated technically with LINQ’s team—was likely to yield better results (better responds to SMEs needs and sector overall). LINQ’s RFA application process is mainly focused on grants and does not provide an illustrative guide of the types of TA that could be provided. As such, it may unintentionally emphasize grant assistance as the main intervention, overshadowing TA. This is also reflected in the fact that grantees do not universally receive TA. Grantees are either provided TA or not, based on either the request of the beneficiary or the recommendation of LINQ’s technical staff. Training is not considered the most strategic and optimal intervention if not linked to grants. A single training session may be insufficient to convince trainees to adopt new production or processing techniques, particularly without sufficient demonstration and support of the SME, cooperative, or association to which they are connected. Furthermore, training that cannot demonstrate impact (in terms of increased sales, lower costs, or increased competitiveness) cannot show an efficient use (cost per impact) of the LINQ activity resources. The current MEL reporting may insufficiently capture beneficiary performance after LINQ assistance due to rapid currency depreciation. While adjusting the MEL reporting indicator for sales mid-performance would be atypical and require additional steps for the MEL manager as well as beneficiaries during reporting, this is not unprecedented: This is similar to what will be done under COVID-19 aligned indicator “Yield of targeted agricultural commodities among program participants with USG assistance,” from the next progress reporting cycle. EQ4: SUSTAINABILITY Which activities or interventions were more sustainable than others, and what were the primary synergies that contributed to that sustainability? 27 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT For the sake of organization, findings for EQ4 are presented per intervention (overall, grants, TA and training, COVID-19/Emergency Response). Overall Grants + TA for Sustainability: Several respondents highlighted the importance of the combination of grants and TA in addressing the needs of agribusinesses and ensuring that the assistance remains sustainable over the longer term. One beneficiary who had received the grant component but not TA yet anticipated that the combination would be important, but could not yet confirm this. Refer to supporting findings under EQ2 and EQ3. Grants Cost-Share Reinforces Sustainability: Beneficiaries were generally committed to sustain the investment made under grants, as reflected in their significant cost share. Maintenance and Operation of Equipment: Beneficiaries generally expressed confidence in their ability to maintain and operate equipment that was supported by grants under LINQ, after the assistance ends. However, several beneficiaries might face challenges in obtaining imported materials needed to operate the equipment, as a result of informal capital controls and the depreciation of the LBP. The explosion at the Port of Beirut on August 4 is likely to increase the cost and time required to import necessary inputs (see Annex 15). Several cooperatives that benefitted from grants noted that the equipment received was already damaged and out of production. Another SME beneficiary is required under the terms of its grant to co￾invest in a piece of equipment, which it is unclear how to put into full operation given the economic situation. One beneficiary (service provider) noted that the quality of materials provided was not excellent, which limited the sustainability of the grant. Environmental Sustainability under Grants: Environmental assessments conducted as part of LINQ’s grant review process extend the review process slightly but ensure environmental sustainability that might otherwise be omitted by agribusinesses given the lack of enforcement of environmental regulations by local authorities. Multiple beneficiaries noted investments oriented to ensure environmental sustainability, particularly around wastewater treatment. TA & Training Environmental Sustainability through Training: One respondent emphasized that training in integrated pest management (IPM) supported environmental sustainability within the agricultural sector by replacing conventional practices reliant on synthetic chemical inputs. Sustainability via TA: Several beneficiaries pointed to the importance of TA in supporting changes in their business operations that staff can sustain over the longer term, particularly through supporting improved profitability and expanding access to export markets. “The combination of the grant and TA is very crucial for any business. And that’s what I like about LINQ. . . It’s not only about getting a machine. It’s about doing things beyond the machine. There is packaging, there is marketing, there is social media, there is an adoption campaign, there is doing tasting in the stores, there is even technical know-how to produce the items . . So the combination of technical assistance and equipment is very important, actually” (Beneficiary-Grantee, Dairy Sector). 28 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Sustainability of Training: Multiple consultants expressed doubts about the sustainability of short￾duration training that is delivered without follow-up support, whether in the form of additional training or continuous local support through partner organizations (e.g., cooperative staff). Some training sessions with planned follow up have been interrupted due to the revolution or COVID￾19, and new schedules cannot be set in light of rising COVID-19 cases. Some farmers who should have attended multiple training sessions are not participating in later sessions, possibly because they fear contracting the virus. COVID-19/Emergency Response Sustainability of Emergency Response: LINQ management acknowledged that the emergency assistance offered in response to COVID-19 (particularly the livelihoods components under Intervention 2 that consisted of materials distribution) is not sustainable, regardless of its applicability to short-term needs. CONCLUSIONS Interruptions to the import of necessary inputs may disrupt the production of several LINQ beneficiaries in the next 12 months. There are clear concerns about the ability of single, stand-alone training sessions to achieve meaningful, long-lasting change within Lebanon’s agri-food sector. It is not clear whether the content of the training delivered by beneficiaries to satisfy a cost-share requirement has been agreed and overseen by LINQ, and whether the trainees have been purposely selected as those best placed to implement the content. A resurgence of COVID-19 may necessitate extended timelines to deliver in-person, field￾based training to farmers/trainees, particularly training that is best delivered in line with agricultural seasonality. EQ5: SCALABILITY To what extent were LINQ’s interventions scalable in terms of size and number? For the sake of organization, findings for EQ5 are presented per intervention (grants, TA and training). Grants Scalability vs. Competitiveness: Some beneficiaries-grantees expressed unease that LINQ should support other businesses in the same sub-sector, particularly within the same district/governorate or targeting the same market niche, as this is perceived to directly introduce new competition for customers or for inputs and therefore undermine their competitiveness. Conversely, other stakeholders appeared more confident that LINQ could support additional beneficiaries without undermining competitiveness of existing beneficiaries. Of these, two grantees expressed a preference that LINQ not extend support to additional beneficiaries in their immediate sector or geography, but confidence that they would remain competitive if this were to happen on the basis of expertise in production and processing. Beneficiaries with access to export markets and/or that 29 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT are stronger in export business expressed less concern if LINQ were to support other, similar businesses due to the fact that export markets are large enough to accept additional supply. Scalability vs. Saturation of Markets: At least one stakeholder noted that LINQ’s targeted sub￾sectors (especially processed foods and dairy) and sub-categories of food within vary in their capacity to absorb new or expanded production. If LINQ scales-up assistance in sub-sectors that are largely saturated, this assistance could undermine the competitiveness of existing beneficiaries. TA & Training Scalability of TA: Several interventions under the umbrella of TA were reported to be time- and resource-intensive, tailored to the specific needs of the business or sector and requiring close collaboration at a person-to-person level with the appropriate staff from the beneficiary business. These activities would be difficult to replicate at scale, without the recruitment of additional consultants. Scalability of Training: Several beneficiaries spoke highly of focused training sessions offered through a workshop or panel presentation format, which allowed multiple beneficiaries (SMEs, cooperatives, etc.) to attend with one or several trainers. Topics mentioned were clustered around food processing or business practices and included labeling for U.S./FDA food export requirements, and branding/social media marketing. Scalability of Training for Traditional Food Processing: Stakeholders expressed conflicting opinions as to whether the market for artisanal, locally-produced processed foods such as traditional mouneh (pantry foods) is saturated or not, which has implications for continued training activities on small-scale food production under LINQ. While one consultant expressed the belief that the market for traditionally produced mouneh products is not saturated in Lebanon and that demand is growing, another consultant took the opposite position and argued that many product categories are saturated. • At least one stakeholder proposed that individual cooperatives would be better to focus their efforts on producing specialized mouneh utilizing products from their surrounding areas. • There are relatively few markets for artisanal, homemade products in Lebanon, and these have been adversely affected by COVID-19 and market closures. CONCLUSIONS LINQ has been limited in its capacity to extend grants, TA, and training assistance on a wide scale due to limited budget resources and a modest performance period. In particular, grant sizes are restricted by limited budget resources as well as the requirement for a cost-share. The scalability of grants is not uniform across sub-sectors. LINQ’s capacity to scale assistance through grant support targeting expanded production within the same region or outside may vary according to sub-sector, sub-category, and the SME’s market orientation (local or export market). LINQ has taken a demand-driven approach to guide its grants and TA, which limits potential for scalability. While there were some limited opportunities to deliver this at scale – e.g., introductory workshops to topics like branding or Food and Drug Administration (FDA) labeling requirements that appear well suited to an initial presentation in a workshop or panel presentation 30 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT format – much of the assistance was necessarily tailored to the exact needs of businesses. This focused and resource-intensive approach inherently limited the scalability of this type of assistance. LINQ’s focus on scalability through training interventions, may have diminished its impact. Conversely, where LINQ tried to aim for scale from the beginning – delivering training to a large number of trainees – it is not clear that the broad approach did much more than reach a large number of people. The evidence is as-yet weak that it delivered meaningful change in their practices, or that it resulted in achievement of target outcomes like increased sales. There is limited information on the real, current opportunity for artisanal mouneh production in Lebanon, particularly in light of market closures and disruptions due to COVID-19. There are differing views about the level of saturation of traditional mouneh products (general or specific products) on the market, which makes it difficult to determine whether there is a viable opportunity to further scale up training to cooperatives for food processing under LINQ. EQ6: ADDITIONALITY To what extent do LINQ’s interventions demonstrate additionality? Evidence of Additionality: Beneficiaries and consultants spoke very positively and strongly about the additionality of LINQ interventions, particularly in light of the challenging economic situation in Lebanon. Some beneficiaries indicated that they would not have been able to undertake the investment in improved production and/or business practices without LINQ’s support. Others indicated that the investment may have happened, but later or on a longer timeline. Evidence of Donor Dependence: Several beneficiaries noted that they had actively sought out the support of donors including USAID in recent years as a mechanism for investment support. Several beneficiaries explicitly noted that they had previously received assistance under other USAID activities, such as Lebanon Industry Value Chain Development (LIVCD) and Lebanon Enterprise Development (LED), and were de-facto referred to the LINQ team for a continuation of assistance. Another beneficiary speculated that follow-on assistance to LINQ could be requested via LED. At least one RFA applicant was, however, denied assistance on the basis of previous support offered under other USAID activities. 12 Beneficiaries-Grantees Avoid Commercial Loans: Several beneficiaries noted that they had not sought out commercial bank loans in recent years, whether because of too-demanding market terms, a 12 This finding was confirmed in an email communication with LINQ management dated August 6, 2020. Q: Would your business have been able to invest in improved production and/or business practices, if it had not received support from LINQ? Why or why not? A: “It would be very hard for us. You have to work more to get more money for the equipment. But [LINQ] is paying half. And for consultants, [LINQ] is also going to pay a lot per hour. So this facilitates a lot” (Beneficiary-Grantee, Dairy Sector). A: “Hardly. Even for the $6,000 co-investment, it was difficult to collect the amount” (Beneficiary-Grantee, Fresh Fruit Sector). A: “At that time, we could not have paid for [the] consultancy services. While [our cooperative] knew that it needed this support, it did not have sufficient cash to cover this TA. We had reasonable confidence in the quality control, but for the costing, we knew we did not know how to do this without expert support” (Beneficiary-TA Only, Food Processing Sector). 31 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT discomfort or dislike of commercial banking institutions, or exclusion from the commercial banking sector (lack of a bank account). Environmental Additionality under Grants: Environmental assessments conducted as part of LINQ’s grant review process extend the review process slightly but are additional in that many beneficiaries may be unable/unwilling to submit to such review without this requirement. A lack of government enforcement of environmental regulations undermines adherence by agro-processors. Additionality of Training: At least one beneficiary-grantee will provide training to farmers as part of the cost-share requirement to receive the assistance from LINQ. The content of these training sessions (good agricultural practices (GAP), improved pruning techniques, IPM) appears to be largely a replication of training that has been given by the beneficiary for years, including previously under other USAID activities. CONCLUSIONS Available evidence suggests that recipients consider USAID/LINQ’s support to be additional. The additionality of LINQ’s support has likely been heightened in the past 12 months, as a result of the increased inability to access investment finance through the commercial banking sector as well as an economic crisis that undermines business performance and thereby ability to self-finance investments. However, evidence of donor shopping and donor dependence undermines evidence of additionality. At least 25% of beneficiaries-grantees interviewed stated that they have obtained or actively seek assistance from multiple donors and/or USAID activities. As for training, training provided by grantees under their cost-share will only support additionality if the training is value-added, reaches a previously unserved population, and/or does not replicate training previously delivered to the trainees under other projects. EQ7: GENDER To what extent were recommendations generated by the gender analysis (USAID Lebanon Gender Assessment13) integrated in LINQ implementation? What are the outcomes generated as a result of LINQ gender mainstreaming? Beneficiary Performance by Gender: LINQ IPTT data through Y2Q3 show wide gaps in performance across gender. Men are overwhelmingly the primary beneficiaries of the activity (Indicator 2). Men also account for the clear majority of individuals enjoying increased incomes as a result of USG assistance (76%) (Indicator 3). Conversely, results from the desk review show that women account for a slightly larger share (59%) of full-time equivalent (FTE) jobs created as a result of USG assistance (Indicator 5), albeit within a small number (9 FTE jobs). While women’s majority in this category initially appears surprising, it may possibly reflect widely reported wage discrimination in Lebanon that depresses women’s wages vis-à-vis those paid to men and perversely incentivizes their hiring over men. Gaps in Gender Mainstreaming: Women’s participation in LINQ activities is targeted through two ways – inclusion of beneficiary screening question around women’s participation within the 13 The USAID Lebanon Gender Assessment was published in January 2019, highlighting the status of gender issues in Lebanon and opportunities to promote women’s empowerment through USAID assistance (including economic growth interventions). Refer to Annex 14 for bibliographic information. 32 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT business/cooperative under the RFA process and the delivery of training to women’s cooperatives. This approach to women’s participation reflects a generally reported bifurcation of women’s and men’s roles within the agri-food sector: • Men work in the field, women work in the packaging/processing/value addition. • Even inside agro-processing facilities, men generally perform roles involving heavy lifting rather than women. • Multiple respondents (particularly consultants) indicated that women tended to exhibit a higher degree of care, attention to detail, and compliance with proper practices such as pruning, trellising, and hygiene and food safety than men. This performance does not appear to be compensated with higher pay but does reinforce the concentration of women’s labor within food processing activities. • One beneficiary specifically noted that the local, rural community actively prevents women from conducting work in the fields. Gendered Wage Gaps: Beneficiaries did not acknowledge that women were paid less than men per hour of work or that women were preferred for processing work for this reason. Limited Messaging on Gender: Overall, beneficiaries and stakeholders generally reported that the LINQ team had provided them with little to no messaging around women’s participation in activities under the project’s umbrella. Exceptionally, a few beneficiaries noted that the LINQ team had emphasized women’s empowerment in the early negotiations of the Business Action Plan with beneficiaries; or that the LINQ team had specifically requested that COVID-19 emergency assistance be delivered to female beneficiaries. CONCLUSIONS LINQ has not consistently integrated the recommendations of the USAID Lebanon Gender Assessment into its implementation, and the lack of gender mainstreaming has correspondingly led to limited outcomes. LINQ does not deliver consistent messaging on the importance of women’s participation to its beneficiaries or stakeholders. The project’s approach to gender mainstreaming has been limited to increased women participation, yet at the same time it reflects a generally reported bifurcation of women’s and men’s roles within the agri-food sector. Training on food safety and food processing techniques in Ainata, North Bekaa in July 2019. A total of 16 women participated in the training. (Source: LINQ) 33 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT EQ8: YOUTH To what extent were youth involved in the LINQ Cooperative Agreement? What can be done to increase youth mainstreaming in LINQ assistance? Lack of Messaging on Youth: Youth were not initially indicated as a target population under LINQ’s design (Cooperative Agreement) or implementation. Consequently, respondents of all types universally reported a lack of messaging around youth participation in activities under LINQ. No tracking of youth was initially required under LINQ’s MEL plan, but recently LINQ adopted a new indicator specific to youth participation: “Youth - % of participants in USG assisted-programs designed to increase access to productive economic resources who are youth (15-29) (IM-level).” Furthermore, LINQ identified a youth focal point to coordinate youth-related issues under the activity. This role had not yet been active during this evaluation’s data collection phase. Youth Beneficiaries: Youth are not a target population under LINQ, and their participation is coincidental rather than a result of any specific targeting or messaging by LINQ. However, youth are reported to be more open to new technologies and management practices than older adults, with respect to both farm practices and food processing techniques. This observation was not universal, with at least one stakeholder noting that youth are not always more open to new practices. This stakeholder instead emphasized the role of education rather than age in opening minds to new approaches. Youth Employment Challenges: Reports have warned14 that (youth) unemployment will rise in Lebanon as a result of existing economic challenges, compounded by the recent explosion at the Port of Beirut. CONCLUSIONS LINQ has had to date no particular focus on youth participation, as this was not explicitly indicated in the terms of its Cooperative Agreement with USAID. Furthermore, LINQ is unlikely to incorporate comprehensive reporting on youth within its performance reports, barring an explicit requirement to do so from USAID. Youth may represent a target sub-population for the introduction of transformative agribusiness models, practices, and technologies in Lebanon. Rising unemployment and fewer job prospects could push Lebanese youth to seek employment and secure income through unfamiliar outlets, including SMEs, cooperatives, or other unknown businesses or in sectors that they might not ordinarily consider as a source of employment. 14 Reuters (2020, August 13). Distribution of seedlings in Mristi, Shouf in April 2020 under the COVID-19/Emergency Response. (Source: LINQ) 34 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT RECOMMENDATIONS EQ1: RELEVANCE How relevant and appropriate are LINQ’s approach and interventions in meeting the objectives and performance indicators for USAID’s Private Sector Development (PSD) Result 115 and Intermediate Result 1.116 and USAID/Lebanon’s Country Development Cooperation Strategy (CDCS) Intermediate Result 2.117? 1. In light of Lebanon’s current economic crisis, specifically following the COVID-19 situation that started in March 2020, LINQ’s interventions should be streamlined to respond to the current and evolving needs of agro-processors (economic/financial and public health) with emphasis on staying within the PSD Result 1 and IR 1.1 and the CDCS IR 2.1. Going forward until the end of the implementation period, the LINQ activity should focus its remaining resources on the grant, TA, and training interventions that are oriented toward the achievement of the PSD and CDCS objectives and limit the re-directing of funds/activities to COVID-19/emergency response (specifically the livelihood￾oriented components under Intervention 2 and Intervention 3). 2. For addressing emergency responses to crises such as COVID-19 or other, a separate mechanism with dedicated human resources should be created within LINQ and/or other USAID projects that can aid, unless the response directly deals with LINQ’s current grantees and TA beneficiaries. EQ2: EFFECTIVENESS To what extent are the LINQ approach, interventions, and results effective in meeting the objectives and performance indicators of the LINQ Cooperative Agreement? What were the outcomes? What were the primary factors responsible for their effectiveness? Management-Level 1. Apply for (LINQ) / Offer (USAID) No-Cost Extension: LINQ should immediately apply for and USAID should approve a no-cost six-month extension, during which LINQ should focus on the implementation and monitoring of grant and TA assistance. Any final performance evaluation may need to treat the last package of grants at the outcome stage solely. For those new grants that may be awarded under the remaining performance period, provisions should be made to extend monitoring and implementation beyond LINQ’s performance period, with the possibility to refer grantees to another USAID activity. 2. Maintain but Reduce Cost-Share: While LINQ should retain a cost-share requirement under its grants, it should discuss reducing the requested proportion with USAID. Ideally, the minimum required cost-share application should be reduced to 25-75 in favor of the beneficiary. For existing beneficiaries, given the worsening economic crisis, LINQ should review its grant agreements with those beneficiaries that demonstrate a serious risk of not meeting the 50-50 cost-share requirement that was imposed at the time the memorandum of understanding (MOU) was signed, to allow a reduction in the cost-share ratio, but without going below the 25-75 level. Moreover, LINQ should 15 PSD Result 1: Private sector competitiveness increased 16 PSD Intermediate Result 1.1: Increased domestic and export sales 17 CDCS Intermediate Result 2.1: Increased domestic and export sales 35 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT review all new grant applications to require the minimum cost-share requirement and award additional points in its criteria review for cost-effectiveness when the cost-share exceeds 25%. 3. Refocus on Original Design and Agro-processors as Entry Point: Comprehensive support packages provided directly to Lebanese farmers – potentially including an array of inputs – should not be pursued further under LINQ. Rather, these should be considered under a forthcoming activity to support the agricultural sector. As for humanitarian/livelihoods support, LINQ should provide USAID/Lebanon with the full list of all beneficiaries that received assistance under COVID￾19/Emergency Response Interventions 2 and 3, for possible follow-on assistance through other implementing activities or mechanisms. 4. Improve Communications: LINQ should emphasize a clearer communication strategy for engagement with beneficiaries and RFA applicants, to ensure prompt, timely, consistent, and correct communications. Moving forward, LINQ should prioritize clear and prompt communication with all RFA applicants – regardless of the final selection status. For those beneficiaries who apply for a grant (equipment/machinery/facilities) but are not awarded one, LINQ should provide clear reasons why it refused their applications and indicate where – if at all – existing but hidden capacity could be maximized. LINQ should work to increase the visibility of both LINQ and USAID as entities within its communications with end-beneficiaries. This may be an area for focus and an opportunity for particular contribution of the CLA team. Operational-Level 5. Focus on Beneficiaries’ Resilience: Moving forward, LINQ should focus on the resilience of existing beneficiaries and work to support their continued operation in the market (at least to maintain sales and employment, if growth is not possible) in light of the multiple challenges facing the economy. For example, within the dairy sector, LINQ should gather its beneficiaries that are goat dairy processors to collectively consider the shortage in goat milk and brainstorm potential solutions to this challenge within their value chain (such as partial or total replacement with cow milk). LINQ should share relevant, value-chain specific findings with other USAID activities that will work with these sectors to inform their future work, as appropriate. LINQ should reach out to cooperatives and their members—possibly through a focus group discussion (FGD)—to assess current/evolving needs and most appropriate interventions to support the continuation of operations or other alternatives (rather than growth) in light of the Lebanese situation and the winter season ahead. 6. Address Input Disruptions: In light of the rising costs of imported raw materials (including packaging materials), LINQ should provide beneficiaries with assistance to identify opportunities to reduce production costs by optimizing the production and selection of lower-cost ingredients and/or modifying their packaging to reduce the amount of materials used. LINQ should compile information on local suppliers (manufacturers or sales agents) of key inputs for which prices have risen rapidly (such as imported packaging materials). For the longer term, USAID should assess within another activity (ARE, TIF, or other) the viability of supporting an existing local manufacturer to expand the supply and/or quality of food-grade packaging materials for use by local agro-processors. 7. Target Market Linkages: In the remaining performance period, LINQ should focus on identifying and fostering market linkages and sales channels for existing beneficiaries that correspond to their comparative advantage and quality-price proposition, and work in tandem to connect those without access to finance to make the most of this channel. LINQ can work with focal points across chambers 36 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT of commerce, syndicates of producers, raw material importers, local material manufacturers, and other stakeholders to foster such linkages. For those agro-processors that cannot reach the export market independently (cannot realistically or practically meet export standards in the near term), LINQ can facilitate market linkages to key domestic food processors that may be willing to buy on an aggregation model, or to mass retail outlets or wholesaler/retailer partners that may be in search of import replacement. LINQ should consider creative options such as virtual matchmaking events to foster linkages. Exporting SMEs should also be trained on financial management (including costing/pricing of products during a period of volatile currency movements) so they could keep on short- and long-term performance of their business. LINQ should provide beneficiaries with TA or training on identifying and targeting a market niche, including on how to make consumers notice their products through association, history, and storytelling. 8. Re-review Grant Applications – Immediately: LINQ should critically, carefully, and quickly re￾review all applications that it received under RFA-001 and/or RFA-002 and has not already rejected as out of scope to ensure that the request remains valid, that it reflects the current economic situation and the applicants’ needs and business opportunities within the current context, and that the proposal meets minimum scores to merit investment by LINQ. The additional re-review step should be immediately communicated to all applicants that have been retained for review, including any requirements for data/information updates and the anticipated timeline for proceeding. Those applications that will be excluded from consideration should be notified of this decision immediately. If the preceding review offers an insufficient pool of applicants to account for the $400,000 of grants, then LINQ should immediately issue a third RFA or an APS to solicit additional proposals. 9. Re-Commit to Demand-Driven Assistance under Grants: For forthcoming grants, LINQ should follow a demand-driven proposal approach and initial consultation process, as was previously done under the RFA process and Business Action Plan negotiation. LINQ should prioritize awarding grants that respond to agro-processors’ actual needs in light of the current circumstances and that recognize the reduction in access to finance due to factors including informal capital controls – even if LINQ therefore falls short of its target to issue 30 grants. Potential needs to consider beyond capital investment may include lower-value grants to support non-capital investment, operating capital to obtain necessary (imported) inputs, and supporting TA to assess pricing and cash management. Applications that will be selected to receive grant assistance under the remaining performance period should be considered for a full package of grant + TA support that can best enable market success within the current economic situation. 10. Target TA and Training Effectiveness: As part of its CLA strategy, LINQ should inform the consultants of the beneficiaries’ feedback on the effectiveness of their support. This might be structured as an ex-post evaluation of the beneficiaries who received TA support through individual consultants. As for training, USAID should validate the effectiveness of the training offered by bringing in external experts to conduct field visits and observations as well as assess the quality of fresh and processed food products, to better assess the effectiveness of this training. LINQ should also ensure that its sampling strategy for the Y2 training survey will include trainees from the dairy sector. LINQ should carefully consider the timing of all remaining training sessions to ensure that training is delivered before the techniques are needed, but not so early as to be forgotten. Under the remaining performance period, follow-up should be provided to address gaps in those trainings already delivered, particularly targeting trainees connected to beneficiaries-grantees. 37 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Strategic-Level 11. Reserve Long-Term Efforts for New Activities: While LINQ should prioritize quick and targeted interventions within the remaining performance period, USAID should support long-term interventions that require import or an extended local manufacturing period for the installation of heavy equipment under its new agri-food activity (ARE). 12. Build Brand Lebanon: Through ARE and/or TIF, USAID should position and promote Lebanese processed foods (including dairy products) as artisanal products on international markets. EQ3: EFFICIENCY To what extent were LINQ’s selected interventions and entry points in each subsector strategic and optimal in achieving the intended results? 1. Clarify Scope of Grants and Emphasize TA under RFA: Under RFA-003/APS, LINQ should continue to conduct outreach events that provide an overview of the proposal components and business case that respond to application requirements. LINQ’s staff should clearly explain to potential beneficiaries the permissible investments under grants as well as systematically consider the capacity of the potential grant applicant (production capacity, management, human resources, infrastructure, etc.) and design and deliver TA packages that help the beneficiary in the short and long term, regardless of whether they receive a grant or not. To have a greater impact through TA, LINQ should make it clear that it can award TA under its RFA process – regardless of whether a grant is awarded or not. The application form should allow the following options: grant / grant + TA / TA only. In addition, LINQ should streamline its TA in a structured way and communicate the types of TA available to the applicants before the RFA deadline. 2. Focus on Priority Needs Post-Crisis: LINQ should discuss with its beneficiaries their priority needs in light of the current Lebanese situation (“nice to have” versus “needs to have”). These needs should be vetted by a technical advisory committee comprised of external technical experts who can verify the most appropriate facility design, production line expansion, etc., and the corresponding procurement decision based on sound business advice. 3. Prioritize Linked Training: LINQ should prioritize training for farmers/processors/individuals linked to grantees or potential grantees, which may offer more immediate opportunities for impact in terms of increased sales or competitiveness compared to independent beneficiaries, and particularly in light of the COVID-19 situation and the impracticality of delivering training to farmers and others through the internet (rural, marginalized without sufficient digital access). Pruning training in Bcharre to support the fresh fruits sub￾sector, April-May 2020. (Source: LINQ) 38 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT 4. Refine MEL Reporting: LINQ should continue to report against the standard PIRS indicator: incremental sales in USD and average price per unit, disaggregated by domestic and export. If the current indicator does not track actual volumes sold at the time of reporting, LINQ should collect and report data on this indicator to account for discrepancies due to rapid depreciation of the LBP. Also, LINQ should require beneficiaries to report on the indicator “number of jobs created/maintained” with a gender disaggregation. (For the findings/conclusions underpinning this recommendation, please refer to the corresponding sections under EQ7.) EQ4: SUSTAINABILITY Which activities or interventions were more sustainable than others, and what were the primary synergies that contributed to that sustainability? 1. Anticipate Supply Disruptions: LINQ should start identifying/documenting the potential disruptions to the grantees’ supply chains as a result of capital controls and/or the inability to import needed materials. LINQ should gather and share information on viable and legal mechanisms to access trade finance to minimize such disruptions. One option to consider is to selectively refer current beneficiaries to the LIFE project, which supports access to finance. 2. Revisit Training under Grant Cost-Share: For those grant/TA beneficiaries that will provide training as part of their cost-share contribution, LINQ should from the outset stipulate the content of the training, the targeted trainees, and the connection of trainees to an agro-processor that can purchase their production. LINQ should consider offering an extension of not more than six months to selected MOUs with beneficiaries that had committed to deliver training as part of their cost￾share, in cases where safe delivery of in-person training sessions is not currently possible in light of COVID-19. 3. Avoid Single, Standalone Training: LINQ should reassess the deployment of single training sessions and begin using them strategically and only when needed. LINQ should critically assess if/where follow-up is needed, preferably linked to a private sector service provider and not just to LINQ’s expert team. EQ5: SCALABILITY To what extent were LINQ’s interventions scalable in terms of size and number? 1. Focus on Impact, Not Scale: Barring any increase in activity budget, LINQ should continue to focus on delivering demand-driven assistance under grants and TA to qualified beneficiaries (especially SMEs) that offer the most promising results (sustainability/sales/financial viability) in terms of effectiveness and efficiency (production/sales/varieties/ technologies). The focus should not be on the scalability of grants or TA under the remaining performance. 2. Deploy Scalable TA and Training Selectively: LINQ should selectively deliver further TA or training that can reach a wider number of beneficiaries where the need is clearest. In some instances, CLA events or workshops on appropriate topics seem well suited: For example, LINQ could present common guidance on the presentation, language, and marketing of export-quality products across a sub-sector for national positioning within targeted export markets (e.g., dairy products in the Gulf countries). Similarly, LINQ should review its menu of TA offerings, identify topics of common interest to multiple beneficiaries in light of evolving market circumstances, and deliver common support. Conversely, before offering additional training on food processing (recipes) to cooperatives, LINQ 39 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT should assess the market opportunity for the specific products and avoid delivering training on the production of products for which the market is saturated or for which the market has (temporarily) ceased to exist. 3. Learn from Firm-Level to Benefit Sector-Level: USAID should look to the experience of LINQ for learning around scalability that could be of use under its newer, larger activities (ARE, TIF). An opportunity here might be to look at lessons from firm-level assistance that can then be built upon at the sector or sub-sector level. By delivering assistance to the sector-level stakeholder (e.g., syndicate or association), LINQ beneficiaries may have an incentive to support the further development of the sector and specifically contribute to building the presence and brand of Lebanese products in the export market. Under a future activity for the agri-food sector, USAID should prioritize common guidance on the presentation and marketing of similar products as national commodities and keep the same language and consideration of quality in outside markets. EQ6: ADDITIONALITY To what extent do LINQ’s interventions demonstrate additionality? 1. LINQ should continue to carefully vet RFA applicants to ensure that beneficiaries are not chronically dependent on donor assistance. LINQ’s additionality can thus be visibility demonstrated and reflected. 2. LINQ’s reporting should specifically indicate those beneficiaries that have received assistance under multiple USAID activities, to the extent that LINQ is aware of it. EQ7: GENDER To what extent were the recommendations generated by the Gender Analysis integrated in LINQ’s implementation? What are the outcomes generated as a result of LINQ gender mainstreaming? 1. The USAID Lebanon Gender Assessment should be a guiding document in LINQ gender strategy and activities; LINQ should identify opportunities to apply the gender analysis’s recommendations across all stages of activity implementation, from design to selection, and make sure their gender strategy is transformative and not simply confirmative of women’s and men’s role in agriculture. For example, going forward: a. LINQ should prioritize outreach to women-owned or women-led SMEs and cooperatives for any further RFA round. b. LINQ should set a 30% target for all further grant awards to reach women-owned or women-led SMEs or cooperatives. c. LINQ (under all future grants to agro-processors) and USAID should consider requiring that beneficiaries implement a productivity or traceability system to track individual performance and productivity at the level of the individual (women and men) and apply that information to ensure fair compensation of women working in agro-processing. LINQ could consider the example of an earlier project implemented by Expertise France in a fresh produce packing unit. d. As part of its CLA efforts, LINQ should develop and share success stories that highlight women and men in agribusiness who have successfully taken on roles outside those typically held by women or men, respectively. 40 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT e. USAID should prioritize marginalized women especially in rural areas under its emergency response to COVID-19 and other economic challenges. EQ8: YOUTH To what extent were youth involved in the LINQ Cooperative Agreement? What can be done to increase youth mainstreaming in LINQ assistance? 1. Under the leadership of its newly identified youth focal point, LINQ should work with USAID to clarify how the project is expected to target youth in its remaining performance period. Examples of this include: a. Encouraging youth SMEs/agro processors to apply for grants; b. Identifying opportunities for increased youth involvement in TA and training; c. Where applicable, reporting on youth within its different indicators. 2. LINQ and USAID (within the context of the forthcoming ARE activity) should identify opportunities where youth in Lebanon may represent a target sub-population for the introduction of transformative agribusiness models, practices, and technologies. LESSONS LEARNED This mid-term performance evaluation highlights the very challenging environment facing Lebanon’s agri￾food sector, as well as the need for continued donor interventions such as LINQ to support the sector and longer-term, real, and equitable economic growth in the country. As multiple crises coincide, it is difficult if not impossible to fully appreciate in the short term the implications of these challenges, or how they can distort firm- and activity-level performance. Nevertheless, flexibility in donor response is important in responding to evolving circumstances, and periodic review is useful to validate and reaffirm the core focus and design of activities such as LINQ. Similarly, an operational-level review is also useful to ensure that interventions best respond to beneficiaries’ current needs and that core support functions such as communications and monitoring are fully mobilized in support of activity objectives. In circumstances such as those now facing Lebanon, efforts to support resilience rather than aggressive growth may be best, even at the expense of meeting performance targets in full. 41 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEXES ANNEX 1. LINQ INCEPTION REPORT [redacted] 42 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 2. EVALUATION DESIGN AND METHODOLOGY This evaluation has relied on a mixed methods approach incorporating both qualitative and quantitative analysis, drawing on both secondary and primary data. A literature review encompassed secondary data, including project documentation (e.g., the contract, work plan, gender analysis, quarterly reports, annual report, and technical reports). Qualitative analysis drew on primary data collected via KIIs with the Mission, LINQ staff, LINQ partners, LINQ beneficiaries, and other stakeholders. Quantitative analysis has drawn on secondary data retrieved from the LINQ monitoring, evaluation, and learning (MEL) system. In accordance with ADS 205.3.8.2, this evaluation has sought to adequately capture the situations and experiences of both males and females and highlight any challenges and/ or achievements in filling gender gaps identified in the USAID Lebanon Gender Assessment and LINQ gender strategy. The evaluation addresses the extent to which the project meets CDCS gender objectives because of the activities that were implemented. DATA COLLECTION The evaluation team used a primarily qualitative approach to answer the evaluation questions, conducting KIIs and group interviews with key stakeholders including beneficiaries (grantees and trainees). In-person data collection has been subject to appropriate risk mitigation measures to account for the COVID-19 outbreak. Interviewees included USAID/Lebanon management, LINQ management and staff, LINQ consultants, and beneficiaries including grantees and end-beneficiaries assisted by LINQ through technical assistance, training interventions, and COVID-19/Emergency Response. The evaluation team has supplemented the qualitative inquiries with a document/desk review, including an examination of quantitative data available from activity monitoring. The evaluation team has not collected primary quantitative data. Data collection began on July 9, 2020 and concluded by July 25, 2020. The KIIs were conducted by the evaluation team in English and in Arabic, upon the preference of the stakeholder. Arabic-language interviews were transcribed and translated into English to facilitate data analysis. The evaluation questions as mapped against data sources, data collection methods, and analysis methods are found in Annex 2. DESK REVIEW The evaluation team conducted a comprehensive literature review of relevant internal and external documents. These included: • USAID/Lebanon’s CDCS • USAID/Lebanon’s Performance Management Plan (PMP) PSD • LINQ’s request for proposal (RFP) • LINQ’s cooperative agreement • Sector assessment carried out by LINQ • LINQ’s annual work plans 43 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT • LINQ’s periodic reports (annual, quarterly, other) • LINQ’s MEL plan • LINQ’s gender strategy • USAID/Lebanon’s Gender Analysis • Supporting, internal documentation deemed relevant to the evaluation questions Documentation and data were requested directly from the LINQ implementing partner. KEY INFORMANT INTERVIEWS KIIs are the evaluation’s primary method and most significant source of data. This method was favored here for two primary reasons: 1) its individualistic nature; and 2) its open-ended/inductive character. The individualistic nature of the KIIs with management (USAID/Lebanon, LINQ management and staff) was important simply because all such managers have unique roles on the project, and those experiences needed to be captured to understand the project’s functionality. However, the individualistic aspect also was pertinent to capture specific experiences and anecdotes among LINQ consultants (business advisors) and trainers, as well as the experience of LINQ beneficiaries (including recipients of grants, TA, training, and COVID-19/Emergency Response assistance). The KIIs’ open-ended and inductive nature was highly relevant because the evaluation team did not enter this evaluation with established theories or hypotheses about what worked and why. We rather sought data that revealed the project’s story and searched for patterns from observation. It is the view of Social Impact and many experts in the field18 that these kinds of inductive approaches are a well￾advised best practice in most instances of retrospective performance evaluations. The KIIs’ sampling and purpose in each respondent category were as follows: • USAID/Lebanon, LINQ Management/Staff: The evaluation team conducted KIIs with a purposive selection of the principal and most relevant managers at USAID/Lebanon (USAID Economic Growth Office, LINQ Contracting Officer’s Representative-COR/Agreement Officer’s Representative-AOR) and LINQ staff (Chief of Party-COP, Technical Manager, Communications & Outreach Manager, MEL Manager, Procurement & Administrative Specialist, Grants Specialist, and Technical Coordinator). The KIIs provided an in-depth understanding of the project’s achievements, the factors that contributed towards achieving the project’s results, recent pivots in the project’s strategy, and implementation successes and challenges. • LINQ Consultants: LINQ provided technical assistance, capacity-building, and training sessions to beneficiaries through two primary channels: (1) recruiting trainers to conduct training sessions; and 2) on-site technical assistance to beneficiaries on specific topics and issues. The evaluation team conducted 14 KIIs with a purposive selection of the consultants who worked under the LINQ activity to deliver technical assistance and/or training to beneficiaries (among a 18 E.g., Guest, G., E. Namey; M. Mitchell (2013). Collecting Qualitative Data: A Field Manual for Applied Research. Los Angeles: Sage Publications. 44 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT total population of 20). A list of consultants and the corresponding number of trainings delivered, and beneficiaries reached was provided by the LINQ implementing partner. Purposive selection targeted the full list of consultants provided by LINQ who reached the largest number of beneficiaries, as they may have a wider view of the activity, its results, and its challenges. The evaluation team conducted KIIs with the consultants that provided technical assistance and training to the different beneficiaries to assess the effectiveness, efficiency, and sustainability of as well as challenges to this intervention. • LINQ Beneficiaries-Grantees: The evaluation team conducted 19 KIIs with a purposive selection of the beneficiaries who applied and received investment grants under LINQ (among a total population of 20). 19 A list of beneficiaries was provided by the LINQ implementing partner. Selection targeted those agribusinesses that were awarded investment grants. The KIIs provided in-depth understanding about the activity’s achievements, factors that contributed towards achievement of the activity’s results, and implementation challenges. • LINQ Beneficiaries-TA Only: The evaluation team conducted 6 KIIs with beneficiaries who received TA only under LINQ (among a total population of 17). A list of beneficiaries was provided by the LINQ implementing partner. A purposive selection targeted those agribusinesses that had applied for investment grants but were awarded TA (among a total of 16), with an effort to reach businesses in each of the three sub-sectors. The KIIs provided in￾depth understanding about the activity’s achievements, factors that contributed towards achievement of the activity’s results, and implementation challenges. • LINQ Beneficiaries-Trainees: The evaluation team conducted 7 KIIs with beneficiaries who participated in training sessions or workshops provided by LINQ (among a total population of more than 1,500). A list of beneficiaries (including names, contact information, and type of training received) was provided by the LINQ implementing partner. The selection of LINQ beneficiaries for participation within the KIIs was purposive, so as to reach beneficiaries receiving different types of training, as well as beneficiaries of distinct socio-demographic profiles (males and females, youth and older adults).20 Because many of the trainings were organized in a specific geographic location, selection for these KIIs was effectively stratified; within each stratum, a random selection of individual beneficiaries was invited to participate in the KIIs. • LINQ Beneficiaries-COVID-19/Emergency Response: The evaluation team conducted 6 KIIs with beneficiaries who received support from LINQ under the COVID-19/Emergency Response, Intervention 2 (among a total population of more than 1,550). A list of beneficiaries (including names and contact information) was provided by the LINQ implementing partner. The selection of LINQ beneficiaries for participation within the KIIs was purposive, so as to reach beneficiaries of distinct socio-demographic profiles (males and females, youth and older adults) across 19 In fact, the total population was only 19, but one additional beneficiary was included in this pool due to an error in the LINQ beneficiary database. In sum, only one beneficiary-grantee was not available for a KII. See Annex 7. 20 For a detailed presentation of the selection of these beneficiaries, refer to Annex 5. 45 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT different regions of Lebanon. 21 Within each stratum, a random selection of individual beneficiaries was invited to participate in the KIIs. The final reach for qualitative data collection differed slightly from what was proposed in the inception report, for reasons including the non-availability of targeted respondents; COVID-19 and related public health restrictions, which prevented the timely conduct of face-to-face data collection; and the decision to include additional beneficiaries were not fully addressed within the inception report. Table 1 summarizes the approach to primary data collection used to support this evaluation. Detailed lists of persons interviewed are presented in Annex 5. KIIs were conducted with beneficiaries located throughout Lebanon, with a large proportion concentrated in major agricultural areas (Akkar, Baalbak￾Hermel, Bekaa, Mount Lebanon, and North). Table 2: Primary Data Collection Approach Stakeholder/Beneficiary Class Target Actual KII Site Visit FGD # Individuals KII Site Visit # Individuals (Gender) USAID 2 4 1 2 (1F / 1M) LINQ Management & Staff 7 7 7 7 (4F / 3M) LINQ Consultants 20 20 14 14 (7F / 7M) LINQ Beneficiaries-Grantees & Grant Applicants 19 3-6 22 19 - 19 (1F / 18M) LINQ Beneficiaries-TA only - - - - 6 6 (1F / 5M) LINQ Beneficiaries-Trainees 4 24-32 7 7 (2F / 5M) LINQ Beneficiaries-COVID-19 Emergency Response - - - - 6 6 (5F / 1M) Other 5 5 - - TOTAL 53 3-6 4 82-90 60 - 61 (21F / 40M) MONITORING DATA ANALYSIS A detailed review of LINQ’s MEL databases and reporting was not originally planned to be included within this mid-term performance evaluation,22 rather the desk review was intended to be limited to key documents such as the MEL plan. However, throughout this evaluation, it was necessary to review LINQ activity databases and indicators in order to extract relevant information such as beneficiary categorization and contact information. In addition, the relevance of performance indicators arose 21 For a detailed presentation of the selection of these beneficiaries, refer to Annex 5. 22 Previous mid-term performance evaluations conducted under PMSPL II have specifically included a detailed MEL review under the theme of “efficiency,” but this level of review was not requested within the corresponding evaluation question for this evaluation (EQ3). 46 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT organically from KIIs with selected stakeholders. As a result, the evaluation team had the opportunity to conduct a partial analysis of the MEL databases and indicators and identify opportunities for improved performance. The evaluation team collected monitoring data from LINQ databases and analyzed them to confirm the progress made toward achieving the project’s goals. In addition, we examined outcome variables at the disaggregated level in as many ways as the data set meaningfully allowed. For example, we considered the length of time in program in relation to the achievement of outcomes and the proportion of accepted grants with respect to the total number of applications (all rounds). ETHICAL & PRACTICAL MEASURES FOR PRIMARY DATA COLLECTION All primary data collection adhered to ethical research standards. A consent form was read prior to each interview, informing participants of the purpose of the evaluation and their engagement with the evaluation team, and of their rights. Verbal consent was collected for KIIs: For data collection conducted by telephone or similar technology, the interviewer obtained verbal consent from the participant and marked the consent form on behalf of the participant. The evaluation team informed participant that notes would be taken and secured participant approval before initiating any audio recording. To ensure data protection and privacy of the audio recordings, the recordings have been kept in the possession of the evaluation team and transferred from the recording device to a password-protected, restricted access data storage location (SharePoint) until the evaluation is complete, at which point the audio recordings will be destroyed. Of particular note, qualitative data collection has been undertaken with an expectation of confidentiality on the part of participants, such that individual responses will not be linked to an individual’s identity, so as to encourage truthfulness in participation. Primary data collection has been conducted remotely (by telephone or digital platform-Microsoft Teams), as an effective means to collect primary data and solicit input despite challenges related to public health and safety, in light of physical distancing requirements linked to the COVID-19 pandemic. Planned, in-person data collection through site visits to conduct KIIs with beneficiaries-grantees and focus group discussions (FGD) with beneficiaries-trainees was not possible to be conducted safely in light of the COVID-19 situation and the appropriately stringent requirements of the Social Impact Institutional Review Board. For this reason, the site visit component was dropped (though remote KIIs were conducted); and the FGD were converted into a KII and conducted using one of the remote technologies referenced above. As a result, fewer beneficiaries-trainees were included within the primary data collection than had been targeted, as KIIs required more time than FGD to reach the same number of individual respondents. Data collection was primarily conducted in English, though Arabic was also used in cases where the participant expressed a preference to speak Arabic. In this case, bilingual members of the evaluation team served as translators for the administration of the KII; notes or transcripts were then translated into English with the support of trained SI staff. Data analysis has been conducted in English. DATA ANALYSIS The evaluation team has conducted qualitative data analysis of the findings and provide conclusions and recommendations accordingly. The evaluation team used a structured and systematic approach to 47 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT analyzing the qualitative data and triangulated multiple methods and sources to ensure the reliability and validity of results. The evaluation team has made use of appropriate software programs to facilitate qualitative data analysis (Dedoose). The evaluation team analyzed the qualitative data as follows: • Recorded notes during or after KIIs, using prepared data collection tools (interview guides), which were reviewed and elaborated as soon as possible following data collection; • Prepared transcripts of audio recordings conducted during KIIs, subject to appropriate consent procedures, as soon as possible following data collection; • Coded KIIs according to themes relevant to the evaluation questions, disaggregating answers for all Key Questions by gender of respondent. When using Dedoose, all typed transcripts for KII were be uploaded into Dedoose, and then Dedoose was used to facilitate coding; • Prepared an evaluation data analysis matrix identifying the themes that emerged in the KIIs to facilitate systematic and rigorous data analysis aimed at identifying key study findings as they relate to the evaluation questions; and • Prepared a detailed outline summarizing key findings based on all the data analysis, conclusions for each study question, and overall recommendations. The evaluation team also conducted quantitative analysis of relevant data, including selected data generated from the MEL system. The evaluation team made use of appropriate software programs to facilitate quantitative data analysis (Excel). Quantitative data analysis was specifically anticipated to be applied to understand the extent of engagement by target or desired socio-demographic groups (sex, age). Anonymized data analysis files developed through this evaluation will be the property of USAID and uploaded on their Development Data Library (DDL). LIMITATIONS & BIASES The evaluation team is aware of several noteworthy biases/limitations in this kind of research: Response Bias: Response bias is the risk that key informants may be motivated to provide responses that would be considered socially desirable or influential in obtaining donor support. The team has mitigated theses risk by minimizing, to the extent possible, the connection between the evaluation activity and any future opportunities for support. The team also reinforced anonymity of respondents during the consent proceedings, which may make respondents less likely to over-report positive data if LINQ will not subsequently know what information was reported by them as individuals. Recall Bias: Recall bias is a common evaluation problem where beneficiaries may respond to questions posed by the evaluation with answers that blend their experiences into a composite memory. Respondents who may have participated previously in similar activities may not have distinguished their experience with LINQ. Additionally, depending on when beneficiaries participated in LINQ activities, their perceptions of events may have changed over time and their ability to remember specific details may fade. The team has sought to mitigate this risk by conducting as many KIIs as possible within the available evaluation period to triangulate responses and increase the validity of the findings. 48 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Selection/Availability Bias: Concerning qualitative data collection, there is a risk that the team might only have reached the most active, responsive, or engaged stakeholders. To mitigate the risk of selection bias, we have relied on random selections in the case of beneficiaries (those beneficiaries that received TA only and trainees) and taken only minimal input from the implementers and partners in this selection. On the monitoring data analysis, there was a comparable risk that the low-performing firms may not have reported full data or may have reported data inaccurately. Gender Bias: Gender bias is a risk because most individuals have a subconscious sense of appropriate roles and behavior for women and men. To ensure the opportunity for female beneficiaries to engage in the evaluation process, a target was set that approximately 30% of beneficiaries participating within the LINQ beneficiary KIIs should be female. The 30% target corresponds to the overall target for female participation in LINQ activities. Of note, KIIs were held with female beneficiaries (trainees, recipients of COVID-19/Emergency Response) to ensure voices of female beneficiaries have been clearly heard and determine whether their experiences with LINQ reflected any significant differences in approach, outcomes, or impacts from those of male beneficiaries. Monitoring Data Bias: The team entered the inquiry under the assumption that up-to-date or nearly up-to-date monitoring data were available for all participating firms, thereby providing quantitative outcome data for the project on a census basis. Limitations Created by Social & Economic Unrest: The ongoing social and economic unrest in Lebanon resulted in some changes to the evaluation plan—such as conducting all interviews by telephone. In addition, many beneficiaries were exceptionally difficult to reach in light of mounting pressures on their businesses, which made scheduling and conducting KIIs a challenge. To mitigate this, the evaluation team persistently followed up with beneficiaries (especially grantees and TA recipients) to schedule (and re-schedule) KIIs at times and dates convenient to them, and focused on most important interview questions to limit the time required to conduct KIIs. For beneficiaries-trainees, the evaluation team over-sampled and conducted KIIs with the beneficiaries who were responsive to the call/invitation; this practical approach may, however, have introduced bias if the responding beneficiaries were systematically different from the non-responsive beneficiaries. In-Person Data Collection: Physical distancing requirements related to mitigating risks of COVID-19 transmission restricted the ability of the evaluation team to conduct in-person data collection. During the initial phase of data collection, the evaluation team prepared and submitted a request for approval to conduct in-person data collection from the Social Impact Institutional Review Board, with an aim to conduct site visits and FGDs later within the data collection period. However, the use of in-person data collection approaches was determined to be infeasible (due to very stringent Social Impact Institutional Review Board requirements for in-person data collection) and ill-advised given a rapid increase in local transmission of COVID-19 during the data collection period. Instead, the evaluation team conducted online data collection to reach the intended participants in FGDs, instead conducting those via remote KIIs (as proposed in the Inception Report). Site visits were similarly deemed inadvisable, and were not conducted. Remote Data Collection: The telephone/internet interview format was a practical option for rapid, intensive data collection in a public health context that limited face-to-face contact during time of 49 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT COVID. This format imposed its own limitations, however: This technology was vulnerable to interruptions, especially during the period of data collection (more frequent power cuts in Lebanon that interrupted phone and internet connections). The use of voice-only calls was appropriate to maintain call quality as well as to require less of stakeholders’ limited data services, but resulted in an inability to see people’s expressions – when uncomfortable, tired, or a new approach to a question may have been needed. DISSEMINATION & UTILIZATION PLAN FINAL RECOMMENDATIONS The evaluation provides a wide variety of actionable recommendations for project adaptations and improvement, as well as a formal recommendation to USAID/Lebanon. The final recommendations highlight whether the strategic shift in LINQ strategy should continue or be changed to respond to SMEs needs within the agriculture and agro-processing sector and whether USAID/Lebanon should continue in similar activities, exploring the challenges and successes that should be taken into consideration if the project scales up. The recommendations also consider the ongoing effects of recent social and political upheaval in the country and how those can and should figure into improvements. We have provided these recommendations to USAID/Lebanon in the form of a post-fieldwork presentation as well as in this report. We remain open to any requests from the Mission to package these recommendations in other formats, such as short research briefs. The team anticipates that the results from the evaluation will be used by USAID/Lebanon during its annual Portfolio Review. Social Impact and its staff will remain available to help advise that process as requested by the Mission. The final evaluation report will become publicly available on the DEC repository. LESSONS LEARNED & ADAPTIVE MANAGEMENT The mid-term evaluation seeks to provide parallel insights to LINQ’s management to adapt and adjust to the implementation challenges identified in the report. Social Impact and the evaluation team have followed USAID/Lebanon’s direction in terms of the extent to which the Mission wants our direct engagement with LINQ and/or other relevant stakeholders. Particularly in relation to Evaluation Question 3 and the project’s M&E practice, the mid-term evaluation has sought to address the extent and effect to which the CLA approach has been incorporated in LINQ. In this way, the evaluation has informed the Mission on how to optimize CLA as part of ongoing project implementation. 50 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 3. GANTT CHART - EVALUATION WORK PLAN Microsoft Excel 97-2003 Worksheet 51 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 4. EVALUATION DESIGN MATRIX 2020-07-07 (LINQ Evaluation Design Matrix).docx 52 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 5 FIELDWORK TOOLS INTERVIEW WITH USAID AND LINQ MANAGEMENT AND STAFF LINQ KII-USAID LINQ Staff_final.docx INTERVIEW WITH LINQ CONSULTANTS LINQ KII-Consultants_final.docx INTERVIEW WITH BENEFICIARIES-GRANTEES LINQ KII Grantees_final.docx INTERVIEW WITH BENEFICIARIES – SMES-TECHNICAL ASSISTANCE LINQ KII SME-Technical Assistance.docx INTERVIEW WITH BENEFICIARIES – TRAINEES LINQ KII Beneficiaries Trainees_final.docx INTERVIEW WITH BENEFICIARIES-COVID-19 EMERGENCY ASSISTANCE (INTERVENTION 2) LINQ KII Beneficiaries-COVID Emergency Response.docx 53 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 6. PERSONS INTERVIEWED Table 3: KIIs with USAID and LINQ Management and Staff Stakeholder Name Title 1. GI USAID Charbel Hanna Carol Brakhya COR Technical Assistant 2. KII Land O’Lakes Celine Abdallah Melki Chief of Party 3. KII Land O’Lakes Roland Boulos Al Khoury Al Andary Technical Manager 4. KII Land O’Lakes Pierre Naim Sawaya Communications & Outreach Manager 5. KII Land O’Lakes Amal Georges El Deek MEL Manager 6. KII Land O’Lakes Monzer Amin Yehya Procurement & Administrative Specialist 7. KII Land O’Lakes Jinane Habib Chalhoub Grants Specialist 8. KII Land O’Lakes Reem Khaled El Derbass Technical Coordinator Table 4: KIIs with Consultants Name of Consultant23 1. KII [redacted] 2. KII [redacted] 3. KII [redacted] 4. KII [redacted] 5. KII [redacted] 6. KII [redacted] 7. KII [redacted] 8. KII [redacted] 9. KII [redacted] 10. KII [redacted] 11. KII [redacted] 12. KII [redacted] 13. KII [redacted] 14. KII [redacted] Table 5: KIIs with Grantee Applicants and Recipient Organizations24 SMEs/Grantees Type of assistance Sector 1. KII [redacted] Facility upgrading Food Processing 2. KII [redacted] Adaptive Management Fresh Fruits 3. KII [redacted] Carob Molasses Production Line Food Processing 4. KII [redacted] Adaptive Management Fresh Fruits 5. KII [redacted] Facility upgrading Food Processing 24 List of SMEs/Grantees was provided by LINQ activity. Liban Village Cold Store was interviewed using the KII tool guide for beneficiary-grantee under the initial understanding that indeed belonged in this beneficiary class. 54 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT SMEs/Grantees Type of assistance Sector 6. KII [redacted] Facility upgrading Food processing 7. KII [redacted] Adaptive Management Fresh Fruits 8. KII [redacted] Facility upgrading Food Processing 9. KII [redacted] Adaptive Management Fresh Fruits 10. KII [redacted] Facility upgrading Food Processing 11. KII [redacted] Non-alcoholic Sparkling Apple Juice Production Line Food Processing 12. KII [redacted] Facility upgrading [Terminated] Dairy 13. KII [redacted] Facility upgrading Dairy 14. KII [redacted] Facility upgrading Dairy 15. KII [redacted] Adaptive Management Fresh Fruits 16. KII [redacted] Applesauce Production Line Food Processing 17. KII [redacted] 18. KII [redacted] Facility upgrading Dairy 19. KII [redacted] Facility upgrading Food Processing Table 6: KIIs with Beneficiaries-Trainees and Beneficiaries – COVID-19/Emergency Response Assistance Type/Category Number of KIIs Gender Location Beneficiaries-Trainees 7 Male (5) Female (2) Training tied to beneficiary-grantee 3 Male (3) Akkar, North Training independent of beneficiary-grantee 2 Male (1) Female (1) Baalbek-Hermel Training workshop 2 Male (2) N/A COVID-19/Emergency Response Beneficiaries 6 Male (1) Female (5) Bekaa, South 13 Beneficiaries of LINQ training were selected by LINQ as beneficiaries in several ways: 1) individuals tied to a beneficiary-grantee or beneficiary-TA recipient who were invited by LINQ to attend training workshops on selected topics; 2) members of an association or cooperative tied to a beneficiary-grantee who received training; or 3) members of an association or cooperative independent of a beneficiary￾grantee who received training. Selection of beneficiaries to participate in KIIs reflected this selection as follows: For the first class, a random selection was made from within the LINQ-provided database of beneficiaries to identify to reach workshop attendees, for which two KIIs were held. For each of the second and third classes, one association/cooperative head was selected randomly from the LINQ￾provided database of beneficiaries and invited to a KII. Thereafter, it was also determined that individual beneficiaries should be reached from within these classes. The selection of individual beneficiaries who received from LINQ either training or emergency assistance (COVID-19/Emergency Response – Intervention 2) was based on their identification as members of cooperatives or associations. Due to time constraints, only a small number of beneficiaries 55 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT could be reached as illustrative of the wider class of beneficiaries. The stratified selection was conducted as follows: The evaluation team selected one organization from each category of beneficiaries - trainees tied to an SME (Basoun), trainees independent of an SME (LOST), and recipients of COVID assistance (Sohmoor), aiming for a geographic spread across regions. The LINQ-provided database of beneficiaries was then sorted according to those three organizations ("Training Location") and date ("Training Date" - oldest to newest). Respondent names were already sorted in alphabetical order. Individuals were selected from the most recent trainings, such that those trainings would be potentially more easily remembered. For Basoun and LOST, the selection was randomized by starting at the bottom per organization and choosing four beneficiaries per organization, selecting every fifth name; the first two to respond to the request for KII were retained. For Sohmoor, the selection similarly started at the bottom of the sorted list and the selection was of four female beneficiaries selecting every fifth female. The variation for Sohmoor was needed to ensure the 30% target female KII respondents, as we proposed in the Inception Report; neither of the other groups appeared to list any female beneficiaries, so this was the only option. 56 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 7. BENEFICIARY SUMMARY AND SELECTION PROCESSES LINQ activity documentation shows that it has provided assistance (grants, TA, and support under the COVID-19/Emergency Response) to a number institutional beneficiaries (SMEs, cooperatives, and associations).25 A total of 56 unique beneficiaries have been identified, as presented in the table below. Table 7: Summary of Beneficiaries Beneficiary Category Number of Beneficiaries Grant 26 19 TA 27 17 TA – COVID-19/Emergency 20 TOTAL 56 LINQ has awarded assistance packages including grants (and technical assistance) to a total of 19 SMEs and cooperatives. Of these, 12 were selected through an RFA process. The remaining seven beneficiaries were identified and selected outside of the RFA process: These include two SMEs representing “low-hanging fruit” that were referred from a previous USAID project and identified as fast-track grants before RFA-001: Agro Cedrus and Les Vergers des Cedres (LVDC). Another five beneficiaries were SMEs or cooperatives identified directly by the LINQ team following the June 2019 hailstorm that caused extensive damage to fruit orchards in north Lebanon: [redacted]. The following table indicates the average grant amount, LINQ contribution, and grantee cost share for all grants awarded to date. The amount requested is also indicated for those beneficiaries that were selected through the RFA process; this figure was higher under RFA-002 than RFA-001, on average. Across all beneficiaries, the average LINQ contribution, grantee cost share, and total grant amount are significantly higher for beneficiaries selected through the RFA process than for beneficiaries selected outside of the RFA process. Notably, the cost-share ratio for beneficiaries selected through the RFA process is approximately 50%, whereas the ratio for beneficiaries selected outside the RFA process is 32%. 25 This analysis is based on a review of LINQ activity documentation, entitled “LINQ_List of Partners_April 2020,” “Database of beneficiaries and Enterprises SI july,” and “Partners data updated August 2020.” Assistance is identified therein as grant, TA (understood to encompass both technical assistance and training formats), and COVID-19/Emergency Response (understood to refer to Intervention 2). 26 Eco Land is identified as a beneficiary-grantee in the file “LINQ_List of Partners_April 2020” but not listed within the file “Database of beneficiaries and Enterprises SI july.” Conversely, Liban Village is not identified as a beneficiary-grantee in the file “LINQ_List of Partners_April 2020” but is listed within the file “Database of beneficiaries and Enterprises SI july.” Including both yields a total figure of 20 beneficiaries-grantees. 27 Zanoubia is identified as a beneficiary of TA/training in the file “LINQ_List of Partners_April 2020” but not listed within the file “Database of beneficiaries and Enterprises SI july.” Including this beneficiary yields a total figure of 17 beneficiaries of TA/training. 57 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Table 8: Summary of Beneficiaries Selected to Receive Grants under LINQ SME / Cooperative Name Sector Selection Mechanism Assistance Provided28 Amount Requested in RFA LINQ Contribution Grantee Cost Share Total Grant Budget Cost￾Share Ratio 1 [redacted] Food Processing RFA-001 Grant $19,000 $15,360 $15,900 $31,260 51% 2 [redacted] Fresh Fruits RFA-001 Grant $90,800 $14,290 $14,400 $28,690 50% 3 [redacted] Food Processing Non-RFA Grant + TA - $78,073 $91,060 $169,133 54% 4 [redacted] Fresh Fruits Non-RFA Grant - $20,500 $3,800 $24,300 16% 5 [redacted] Food Processing RFA-002 Grant + TA $109,500 $97,450 $90,300 $187,750 48% 6 [redacted] Food Processing RFA-001 Grant + TA $50,000 $83,000 $114,350 $197,350 58% 7 [redacted] Fresh Fruits Non-RFA Grant - $24,600 $6,000 $30,600 20% 8 [redacted] Food Processing RFA-001 Grant +TA $40,000 $39,500 $41,875 $81,375 51% 9 [redacted] Fresh Fruits Non-RFA Grant - $20,925 $6,000 $26,925 22% 10 [redacted] Food Processing RFA-002 Grant + TA $86,100 $83,000 $114,350 $197,350 58% 11 [redacted] Food Processing RFA-002 Grant + TA $30,350 $45,700 $31,250 $76,950 41% 12 [redacted] Dairy RFA-001 Grant + TA $44,700 $67,650 $69,170 $136,820 51% 13 [redacted] Dairy RFA-002 Grant +TA $20,550 $47,003 $35,800 $82,803 43% 14 [redacted] Dairy RFA-001 Grant + TA $85,000 $99,100 $104,500 $203,600 51% 15 [redacted] Fresh Fruits Non-RFA Grant - $20,205 $6,000 $26,205 23% 16 [redacted] Food Processing Non-RFA Grant + TA - $12,765 $20,360 $33,125 61% 17 [redacted] Dairy RFA-001 Grant + TA $100,000 $91,200 $99,500 $190,700 52% 18 [redacted] Food Processing RFA-002 Grant + TA $109,913 $99,200 $175,000 $274,200 64% 19 [redacted] Fresh Fruits Non-RFA Grant - $22,800 $8,700 $31,500 28% RFA-001 $61,357.14 $58,585.71 $65,670.71 $124,256.43 52% RFA-002 $71,282.60 $74,470.60 $83,270.00 $157.740.60 49% Non-RFA Selection - $28,552.57 $20,274.29 $48,826.86 32% Total $65,492.75 $51,701.11 $53,577.11 $105,278.21 44% 28 Information in this column is derived from KII responses and updated based on the data file “LINQ RFA details_Aug.2020” provided by LINQ management. 58 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT A total of 17 SMEs and cooperatives were selected to receive assistance in the form of technical assistance or training only. Of these, six were identified through the RFA process. One beneficiary that had intended to apply for assistance through the RFA process but exceeded the application deadline, was retained for selection outside of the RFA process. The selection mechanism for the remaining nine beneficiaries is not clearly indicated in the available LINQ activity documentation.29 Table 9: Summary of Beneficiaries Receiving Technical Assistance under LINQ SME / Cooperative Name Sector Selection Mechanism Assistance Provided Amount Requested in RFA LINQ Contribution 1 [redacted] Food Processing RFA-001 TA $39,200 $2,900 2 [redacted] Food Processing Non-RFA TA $3,500 3 [redacted] Fresh Fruit Non-RFA TA $1,500 4 [redacted] Fresh Fruit Non-RFA TA Volunteer 5 [redacted] Food Processing RFA-001 TA $30,000 $1,500 6 [redacted] Food Processing Non-RFA TA $1,500 7 [redacted] Food Processing Non-RFA TA $1,500 8 [redacted] Dairy Non-RFA TA $3,600 9 [redacted] Fresh Fruit Non-RFA TA 30 $4,000 10 [redacted] Dairy RFA-001 TA $49,550 $2,700 11 [redacted] Food Processing RFA-001 TA $58,987 $7,300 12 [redacted] Fresh Fruit and Food Processing Non-RFA TA $3,550 13 [redacted] Food Processing Non-RFA TA $1,500 14 [redacted] Food Processing RFA-001 TA $1,500 15 [redacted] Food Processing Non-RFA TA $4,500 16 [redacted] Food Processing RFA-002 TA $15,207 $1,500 17 [redacted] Fresh Fruit Non-RFA Selection TA $2,000 Total $2,784.38 A total of 20 SMEs, cooperatives, and associations were selected to receive assistance under the COVID-19/Emergency Response (Intervention 2) only.31 These beneficiaries were identified directly by LINQ technical staff, seeking out reliable entities operating on the ground that could distribute the 29 LINQ management clarified that, “Partners from outside the regular process reach out to LINQ team directly or through the Application for Technical Assistance that has been publicly posted. Selection is always based on the same criteria and the LINQ technical team proceeds to a gap analysis before deciding on any type of assistance” (email communication dated August 6, 2020). 30 Jibal Loubnan is identified as a beneficiary of TA/training in the files “LINQ_List of Partners_April 2020” And “Partners data updated August 2020.” However, a respondent indicated that this beneficiary had been disqualified from receiving assistance. 31 Beneficiaries receiving grants and/or technical assistance also received support under the COVID-19/Emergency Response (Intervention 1). At least one beneficiary receiving grant assistance specifically noted that it had also played a role in distribution of assistance under COVID-19/Emergency Response (Intervention 2). 59 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT seedlings and potato seeds to local farmers; and ensure continuity post-distribution to raise awareness and deliver (limited) technical assistance to ensure recipients know how to grow the seeds/seedlings; and collect the results at the later stage. LINQ management further justified the selection of beneficiary organizations outside LINQ’s grantees and TA recipients as follows: “As this was an emergency response that aimed at providing rapid support to small farmers and households, which normally do not constitute the profile of beneficiaries under the LINQ project (the main partners are usually the agribusinesses), LINQ had to look for new adequate partners that are directly linked to a number of farmers. Moreover, LINQ needed to have a wide geographic coverage to reach people across Lebanon” (email communication of August 6, 2020). Table 10: Summary of Beneficiaries Receiving COVID-19/Emergency Response Assistance (Intervention 2) under LINQ SME / Cooperative Name Assistance Provided 1 [redacted] Seeds/Seedlings Distribution 2 [redacted] Seeds/Seedlings Distribution 3 [redacted] Seeds/Seedlings Distribution 4 [redacted] Seeds/Seedlings Distribution 5 [redacted] Seeds/Seedlings Distribution 6 [redacted] Seeds/Seedlings Distribution 7 [redacted] Seeds/Seedlings Distribution 8 [redacted] Seeds/Seedlings Distribution 9 [redacted] Seeds/Seedlings Distribution 10 [redacted] Seeds/Seedlings Distribution 11 [redacted] Seeds/Seedlings Distribution 12 [redacted] Seeds/Seedlings Distribution 13 [redacted] Seeds/Seedlings Distribution 14 [redacted] Seeds/Seedlings Distribution 15 [redacted] Seeds/Seedlings Distribution 16 [redacted] Seeds/Seedlings Distribution 17 [redacted] Seeds/Seedlings Distribution 18 [redacted] Seeds/Seedlings Distribution 19 [redacted] Seeds/Seedlings Distribution 20 [redacted] Seeds/Seedlings Distribution 60 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 8. RFA PROCESS AND SELECTION RESULTS 32 LINQ solicited applications from SMEs, cooperatives, and associations seeking assistance in the form of grants and technical assistance through two Requests for Applications (RFAs). The two RFAs were launched in January 2019 and August 2019, respectively. LINQ received a total of 48 applications in response to its first RFA, and 25 applications in response to its second RFA. The LINQ team screened applications using an evaluation matrix containing multiple criteria to assess compliance with LINQ’s objectives and requirements. These criteria were amended between RFA-001 and RFA-002 as reflected in the following table. The reason for the revision to the RFA criteria was presumably a result of LINQ’s internal learning process: LINQ management streamlined the review process between RFA-001 and RFA-002 to omit certain steps at the application level and reduce the time to move from application to implementation. In addition, the criterion “Viability and Sustainability” did not stipulate a minimum cost-share level under RFA-001 (which had been noted as 30-50% cost￾share within the RFA announcement); however, the criterion was noted to require “a matching co￾investment of at least 1:1” under RFA-002.33 Table 11: RFA Evaluation Criteria RFA-001 RFA-002 Evaluation Criterion 1: Potential Impact 20 30 Evaluation Criterion 2: Viability & Sustainability 20 20 Evaluation Criterion 3: Feasibility 20 20 Evaluation Criterion 4: Additionality 10 10 Evaluation Criterion 5: Scalability 10 - Evaluation Criterion 6: Women’s Inclusion & Promotion 10 10 Evaluation Criterion 7: Employment Created 5 10 Evaluation Criterion 8: Environmental Impact 5 - Total Points 100 100 Table 12: Summary Statistics for RFA-001 and RFA-002 RFA-001 RFA-002 Total number of applications 48 25 Number of applications that “pass” for pre-selection (≥60) 20 14 Of which, number of applications selected for grant 7 6 Number of applications that “fail” (<60) 28 11 Of which, number of applications that are “out of scope” (=0) 18 3 “Pass” rate (pass/total) 42% 56% Average funding request $75,301.56 $76,802.97 Average funding request among applications that “pass” $70,104.95 $64,377.00 32 Analysis is based on a review of the following files: “LINQ RFA_001-V2,” “RFA-001 Subaward Selection Matrix,” “LINQ RFA_002_Aug.19,” and “RFA-002 Subaward Selection Matrix.” 33 The reason for the effective increase in the cost-share requirement was not raised by LINQ management during the KII process. Curiously, data analysis shows that the final cost-share was slightly higher under RFA-001 than RFA-002, despite this change (see Annex 7). 61 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Activity documentation indicates that several proposed investments that failed to meet the minimum evaluation criteria for pre-selection under LINQ were retained for consideration for technical assistance only or referred to another USAID activity (Farmer-to-Farmer). Under RFA-001, applications that were pre-selected scored best in terms of viability and sustainability (average score 14.7 of 20 or 74%), and worst on additionality (average score 6.4 of 10 or 64%). Under RFA-002, applications that were pre-selected scored best in terms of potential impact (average score 23.1 of 30 or 77%), and worst on employment created (average score 6.8 of 10 or 68%). Under RFA-001, applications that were selected to receive grants scored best in terms of viability and sustainability (average score 15.0 of 20 or 75%), and worst on environmental impact (average score 3.1 of 5 or 63%). Under RFA-002, applications that were selected to receive grants scored best in terms of vitality and sustainability (average score 16.2 of 20 or 81%), and worst on employment created (average score 7.2 of 10 or 72%). Through the RFA process, LINQ selected a total of 13 applications to receive grants and technical assistance; another six were selected for technical assistance only. Table 13: Summary of Successful RFA Applications SME / Cooperative Name Selection Mechanism Assistance Provided 1 [redacted] RFA-001 Grant 2 [redacted] RFA-001 TA 3 [redacted] RFA-001 Grant 4 [redacted] RFA-001 Grant + TA 5 [redacted] RFA-002 Grant + TA 6 [redacted] RFA-001 Grant + TA 7 [redacted] RFA-002 Grant + TA 8 [redacted] RFA-002 Grant + TA 9 [redacted] RFA-001 Grant + TA 10 [redacted] RFA-001 TA 11 [redacted] RFA-002 Grant + TA 12 [redacted] RFA-001 Grant + TA 13 [redacted] RFA-001 TA 14 [redacted] RFA-001 TA 15 [redacted] RFA-002 34 - 16 [redacted] RFA-001 Grant + TA 17 [redacted] RFA-001 TA 18 [redacted] RFA-002 Grant + TA 19 [redacted] RFA-002 TA 34 Applied under both RFA-001 and RFA-002. Application under RFA-001 indicated as incomplete. Initial selection was later rejected by USAID/Lebanon (email communication from LINQ management dated August 6, 2020). 62 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 9. WAY FORWARD – RECRUITMENT AND SELECTION OPTIONS FOR FUTURE GRANTS & TA Within the remaining performance period, LINQ plans to distribute approximately $400,000 in remaining budget to grant and TA recipients. An RFA process was used to identify most but not all recipients of the grants issued to date (see Annex 7). The LINQ management team has noted multiple options to recruit and select beneficiaries for remaining grants and TA. This annex attempts to systematically consider the relative merits and limits of the three options that have been raised within the evaluation process: selection of beneficiaries from applications received under RFA-001 and/or RFA￾002, selection of beneficiaries from applications to be received under an RFA-003, or selection of beneficiaries from applications to be received under an APS. OPTION 1: SELECTION OF BENEFICIARIES FROM APPLICATIONS RECEIVED UNDER RFA￾001 AND/OR RFA-002 During the course of qualitative data collection, some LINQ management explained that a planned third RFA may not be conducted. Instead, remaining activity funds for grants and TA may be awarded to applicants that previously passed the screening in earlier RFA rounds. Outstanding Considerations: It is unclear whether LINQ will make its selection from either RFA-001 or RFA-002 or both; whether its selection will be on the basis of the raw scores achieved during the evaluation process; and, if so, whether the previously selected applicants from RFA-001 and RFA-002 will be re-screened using a single, consistent set of criteria (presumably those for RFA-002, which include the 1:1 cost-share requirement) for the sake of fairness in evaluation. Advantages: • LINQ management and staff resources would not be needed for the design and release of another RFA and could instead focus on the review and selection of applications that have already been subject to pre-screening. • Applications received under RFA-001 and RFA-002 that were pre-selected and remain under consideration, but which have not yet benefitted from grants, have requested assistance that exceeds the remaining grants budget (see Table ). In this case, selection from this pool may be more than sufficient to expend activity funds and represent a sound use of management and staff resources. Table 14: Pre-Selected Applications Not Yet Awarded Grants RFA-001 RFA-002 Total Number of Applications 13 8 21 Funding Amount Requested (cumulative) $1,042,704 $529,666 $1,572,370 63 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Disadvantages: • Applications already received under RFA-001 and RFA-002 may no longer accurately reflect those agro-processors’ current situation, most pressing business needs, and capacity for co￾investment in light of the evolving economic situation. • Selection from applications received under RFA-001 and especially RFA-002 may unnecessarily exclude promising agro-processors that chose not to apply, because they would have been unable to meet the 30-50% or 50% cost-share requirement previously imposed. • In the event that the cost-share requirement is relaxed (as recommended elsewhere in this report), then the selection from applications received under RFA-001 and RFA-002 may be seen unfavorably by those beneficiaries who applied under these rounds, were selected for grants, and were required to abide by the earlier cost-share requirement. In this case, LINQ might anticipate complaints and prepare an adequate response if such issues are raised by beneficiaries. Recommendations if Selected: • LINQ should critically, carefully, and quickly re-review all applications received under RFA-001 and/or RFA-002 that have not already been rejected as out of scope and that remain in the ‘pre￾selected’ category to ensure that the request remains valid, that it reflects the current economic situation and applicants’ needs and business opportunities within the current context, and that the proposal meets minimum scores to merit investment by LINQ. The additional re-review step may include a re-scoring of the application against the criteria/scoring sheet used for RFA￾002. The additional re-review step should be immediately communicated to all applicants that have been retained for review, including any requirements for data/information updates and the anticipated timeline for proceeding. • If the preceding review offers an insufficient pool of applicants to account for the $400,000 of grants, then LINQ should immediately issue a third RFA to solicit additional proposals. OPTION 2: SELECTION OF BENEFICIARIES FROM APPLICATIONS TO BE RECEIVED UNDER RFA-003 During the course of qualitative data collection, some LINQ management noted that a third RFA is planned for late 2020. Outstanding Considerations: • None yet identified. Advantages: • An RFA may be preferable was preferable to an APS to facilitate management, as it allows LINQ staff to bundle and review applications along a consolidated timeframe. • Applications received under RFA-001 and RFA-002 that were pre-selected, but which have not yet benefitted from grant assistance, could be invited to re-apply under RFA-003 with an opportunity to update the application to reflect their evolving business circumstances and needs.35 35 NOTE: RFA-001 and RFA-002 explicitly state that, “Applicants that receive technical assistance only under this RFA are eligible to re-apply for investment grants under subsequent RFAs.” 64 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT • An RFA process offers opportunities for co-design, and can be tailored to the requirements of the partner. Disadvantages: • A new RFA will require an outreach campaign to solicit applications. An RFA imposes significant pressure on LINQ staff to reach out to the maximum number of potential beneficiaries in a limited period of time. • An RFA process increases the chances for competition between MSMEs and a prioritization of applications, though this should not be the intention (refer to RFA evaluation criteria 1 and 2) nor the result (e.g., RFA-001 and RFA-002 established a ceiling on the number of grants to be awarded – 8-10 under RFA-001 and 20 under RFA-002). • A new RFA may be less favorable to those applications that were previously received under RFA-001 and RFA-002, as its formal review and selection structure may unintentionally assess applications in comparison to other applications (relative selection, above and beyond the minimum selection criteria). Communication with the existing applicants would need to be carefully managed to ensure that they are clear on the re-application process and that their reapplication is not a guarantee of selection. OPTION 3: SELECTION OF BENEFICIARIES FROM APPLICATIONS TO BE RECEIVED UNDER APS During the course of qualitative data collection, a beneficiary raised the possibility to use an APS mechanism to solicit applicants for grants under the LINQ activity (U-1). Outstanding Considerations: • It is unclear whether an APS can be issued with an understanding that applications received after a set deadline would need to be shifted to another USAID activity for implementation; and whether USAID would accept to impose this shift/referral on that activity. Insofar as LINQ was always envisioned as a “bridge” activity, this assumption seems plausible but should be verified. Advantages: • An APS process offers opportunities for co-design and can be tailored to the requirements of the partner. • An APS would offer a wider time window for beneficiaries’ submissions and participations, as such LINQ would not be closing its door in front of potential beneficiaries because it has reached its obligated ceiling/capacity or because the one month 10 days submission period has ended. The APS can be managed in a way to ensure continuous enrollment while ending one year before the project end date (currently end-November 2021) to allow time for proper implementation. • An APS approach may be preferable to an RFA for the flexibility in reaching potential beneficiaries in the agri-food sector that have had no or limited interaction with USAID or other donors, given its wider timeframe. • An APS approach may be preferable to an RFA for the flexibility in reaching beneficiaries on a time cycle that better reflects their needs (e.g., applications submitted in line with their seasonality of production). • Applications received under RFA-001 and RFA-002 that were pre-selected, but which have not yet benefitted from grant assistance, could be invited to re-apply under an APS with an 65 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT opportunity to update the application to reflect their evolving business circumstances and needs.36 • An APS may be favorable to those applications that were previously received under RFA-001 and RFA-002, as its rolling review and selection structure assesses applications against set criteria (minimum selection) rather than in comparison to other applications (relative selection, above and beyond the minimum selection). Well-motivated applicants able to deliver a revised application quickly would be better positioned to move quickly. Nevertheless, communication with these applicants would need to be carefully managed to ensure that they are clear on the re-application process and that their reapplication is not a guarantee of selection. Disadvantages: • Given the limited performance period remaining, an APS will require an outreach campaign to solicit a sufficient number of applications to utilize the remaining grant budget. In the current circumstances, an APS would impose significant pressure on LINQ staff to reach out to the maximum number of potential beneficiaries in a limited period of time. • An APS would require LINQ staff to manage applications without the advantage of a single timeline to structure review and communication with beneficiaries. GENERAL RECOMMENDATIONS • Recommendation: With limited time and resources remaining, LINQ should continue to focus its efforts on the sectors in which it has previously worked and therefore where its team has developed the greatest focus – the processed food sector (including dairy) and the fruit sector. Accordingly, under RFA-003 or an APS, LINQ should amend the recruitment announcement to remove the following language: “However, the project is interested in truly innovative and promising ideas. Thus, applications from all agricultural sub-sectors are welcome and will be considered.” In the event that LINQ receives expressions of interest from potential applicants that are outside the sectoral focus, they should be referred to the forthcoming agricultural project (ARE). • Recommendation: In light of the difficult economic context and the extremely challenging financial sector/access to finance situation facing the agri-food sector, LINQ should revisit its cost-share requirement to be more favorable to beneficiaries while maintaining a minimum requirement for cost-effectiveness – regardless of which of the three recruitment and selection mechanisms it will apply. 36 Ibid. 66 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 10. BENEFICIARY-TRAINEE SURVEY PROTOCOL AND RESULTS 37 LINQ has conducted ex-post surveys of beneficiaries who participated in training sessions, as a means to document results and track the progress of its activities, specifically against the indicator “Number or farmers or microenterprises who have applied new technologies or management practices as a result of USG assistance.” LINQ administered the surveys by telephone. SURVEY PROTOCOL The LINQ MEL team has prepared a survey form to collect information in a consistent fashion, encompassing open-ended and close-ended questions. The survey was designed to reach trainees including both farmers and individuals who conduct food processing activities. The survey form collects basic information under a section entitled “Section II: Survey Identifier Information.” This section appears to refer to the location of the training, as opposed to the residence of the trainee. • Finding: The form omits the governorate of Nabatiyeh. >> Conclusion: The form does not include all governorates as selection options. >> Recommendation: Add as an option under “Governorate” the eighth governorate of Lebanon, Nabatiyeh. The survey form collects general and specific information under a section entitled “Section III: Respondent Information.” • Finding: The tick boxes for “Age or Year of Birth” are incomplete in that they omit the age of 30. >> Conclusion: The form does not include all ages as selection options. >> Recommendation: Refine the options under “Age or Year of Birth” to ≤ 30 and 30 +. • Finding: The sub-section “B. Specific Information” does not provide an option to solicit information on animal farmers or individuals practicing animal husbandry. >> Conclusion: This omission may be inappropriate, as dairy is one of the three sub-sectors targeted under LINQ. >> Recommendation: The form should be validated against the types of trainings given, to determine if an option for responses around animal husbandry are needed. • Finding: The sub-section “B. Specific Information” allows respondents to indicate whether they have attended one or more trainings (“Check all that applies [sic]”), but only allows a single response in response to the question “2. Did you find the trainings useful?” >> Conclusion: For respondents who attended more than one training – which represents the majority of respondents, as indicated below – feedback on the usefulness of the training is therefore aggregated and not specific to individual trainings. This aggregation could obscure potentially important findings, and limit the utility of the survey results. >> Recommendation: The form should be amended to allow respondents to provide an assessment of the usefulness of trainings, per training. SURVEY RESULTS The LINQ team has provided the results of a survey conducted in mid-December 2019 that targeted 202 trainees (farmers or food processors), from a total population of 424 trainees who were reached from the beginning of LINQ activity implementation through Y1/Q4. The survey reached a total of 204 37 This analysis draws on information included in the documents titled “Farmer Survey Final template En” and “LINQ Y2Q1 survey-summary of results” as well as information provided within relevant KIIs. 67 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT trainees, which exceeds the target sample size. The sampling strategy relied on simple random sampling across the entire population without stratification; therefore, no adjustment is needed to interpret the survey results as representative of the wider trainee population. Trainees are overwhelmingly male (86%), with a minor proportion female (14%). • Finding: Based on the farmer/trainee survey results as of Y1Q4, the proportion of female trainees reached falls below LINQ’s target for gender inclusion, for not less than 30% of beneficiaries to be female. The distribution of trainees by governorate is highly concentrated in two governorates of Lebanon: North (46% + 4% = 50%)38 and Baalbek-Hermel (43%). The remaining trainees are located in Akkar (3%) and Mount Lebanon (3%). No trainees are reported in Beirut, Bekaa, Mount Lebanon, Nabatiyeh, or South. In this way, LINQ’s trainings may be seen to target scale within a limited geographic scope. The most prevalent type of training delivered was on GAP-General (81%), followed by IPM (59%) and food safety and food processing (14%). More than half of trainees benefitted from more than one type of training (57%), most commonly a combination of trainings on crop management (GAP + IPM) (56%). Trainees overwhelmingly reported that trainings were useful “to a great extent” (94%). Remarkably, only a single respondent (0.5%) found the training to be “not at all” useful. These very strong results speak favorably of LINQ’s training activities. A clear majority of trainees (77%) reported that they applied new technologies or management practices as a result of the training(s). Disaggregated by gender, a slightly higher proportion of men (90%) than women (86%) reported that they applied new technologies or management practices as a result of the training(s). The types of technologies or management practices applied by trainees reveal stark differences, as tabulated below. If taken at face value, the GAP-General and food safety and processing trainings have the best rates of application, at 79% among trainees who received this type of training (131 of 166 trainees and 23 of 29 trainees, respectively). The results of this question are not disaggregated by either gender or by age of respondents, despite the fact that multiple consultants/trainers observed that women are more attentive and detail-oriented in training settings and that younger adults are more willing than older adults to adopt new technologies and management practices. • Finding: Despite the option to indicate more than one technology or management practice applied, the survey results report only a single response per trainee. >> Conclusion: The survey form as currently applied may omit important findings among those trainees who benefitted from more than one type of training. >> Recommendation: Allow or require survey respondents who completed more than one type of training to answer whether they have applied more than one new technology or management practice. • Finding: Survey results for Section III, Sub-Section B, Question 4 have not been disaggregated by gender. >> Conclusion: Additional analysis of this question by gender may be merited based on the observation of multiple consultants/trainers that women are more attentive and detail-oriented in 38 Results are presented in the form of a bar graph, two of the bars are labeled “North.” 68 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT training settings. >> Recommendation: Conduct a further layer of analysis of this question with respect to gender of respondent. • Finding: Survey results for Section III, Sub-Section B, Question 4 have not been disaggregated by age. >> Conclusion: Additional analysis of this question by age may be merited based on the observation of multiple consultants/trainers that younger adults are more willing than older adults to adopt new technologies and management practices. >> Recommendation: Conduct a further layer of analysis of this question with respect to age of respondent. Table 15: Farmer/Trainee Survey Results Trainees 39 Trainees Applying a New Technology or Management Practice # % (of 204) # % (of 157) GAP – General 166 81% 131 83% Of which, Pruning 112 71% Of which, Plowing 4 3% Of which, Fertilization 9 6% Of which, Irrigation 6 4% GAP – Olives 7 3% - - IPM 120 59% 3 2% Of which, Application of Insecticides 2 1% Of which, Application of Fungicides - - Of which, Spraying Dates 1 < 1% Of which, Spraying Techniques - - Food Safety & Food Processing 29 14% 23 11% Of which, Food Safety 17 8% Of which, Jam (processing) 2 1% Of which, Tomatoes (processing) 4 2% Total 204 157 Among those trainees who applied a new technology or management practice, an overwhelming proportion indicated that they were greatly satisfied (95%) or somewhat satisfied (5%) with the new technology or practice. • Finding: Despite the option to indicate more than one technology or management practice applied, the survey results report only a single response per trainee to the question on degree of satisfaction. >> Conclusion: The survey form as currently applied may omit important findings among those trainees who benefitted from more than one type of training. >> Recommendation: 39 Figures will not sum to 204 or 100%, as multiple responses were permitted per respondent. 69 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Allow or require survey respondents who completed more than one type of training to answer the degree of satisfaction with more than one new technology or management practice. Among those trainees who have not applied a new technology or management practice (N = 47), 32 trainees offered substantive responses to explain why they have not yet done so. These responses were clustered as follows: the trainee did not yet start his/her work in the orchard or field to apply the technique (N=9); the trainee had already completed the work for the season, prior to the training (N=6); the trainee did not work in the orchard or field this season/year (N=8); the trainee lacked water resources to apply the new techniques (N=3); the trainee already knew the training practices (N=3); the trainee was not interested to change practices (N=1); or the trainee did not prepare mouneh after the training date (N=2). • Finding: Among those trainees who had not implemented the techniques or practices covered in the training, and who offered a substantive explanation as to why, the largest proportion (17 of 32) indicated that the timing of the training was the reason. >> Conclusion: Farmer/trainee responses suggest the greatest challenge to implementation of practices covered in training sessions is the timing of the session. >> Conclusion: Carefully consider the timing of training sessions to ensure that trainings are delivered before the techniques are needed, but not so early as to be forgotten. • Finding: Among those trainees who had not implemented the techniques or practices covered in the training, and who offered a substantive explanation as to why, none indicated that it was because they were unsure of the technique or practice or that they wished for additional follow-up from the trainer. 70 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 11. MEL-INDICATOR PERFORMANCE TRACKING TABLE (IPTT) ANALYSIS 40 Within its IPTT, LINQ monitors activity performance across 17 indicators (Table 14). • LINQ Management and Logistics >> Finding: LINQ has proposed to monitor annually Indicator 6: Macroeconomic / financial situation developing in Lebanon conducive for investment, through a narrative analysis within its annual reports to USAID/Lebanon. No further information on this indicator is included within the IPTT data provided, and no explanatory note is provided. • LINQ Management and Logistics >> Finding: IPTT Indicator 7: Value of incremental sales (domestic and export) collected at firm level for enterprises as a result of USG assistance should be reported in USD (according to the Y2Q2 file units notation), but the IPTT data for Y2Q3 are displayed in % terms. The reason for this mismatch is unclear. The same issue applies to Indicator 8. • LINQ Management and Logistics >> Finding: LINQ has proposed to monitor Indicator 16: Subsidized loans accessed through banks and financial institutions, through secondary data collection from Kafalat on a quarterly basis as well as through a narrative analysis within its quarterly and annual reports to USAID/Lebanon. No data is reported against this indicator within the IPTT provided. IPTT data through Y2Q3 of LINQ’s performance period (through June 2020) indicate that the activity has not yet achieved in full any of its LOP performance targets excepting for Indicator 4, which is a relative target (not less than 30% of all beneficiaries should be female) rather than an absolute target (Table 14). Accordingly, it does not appear that any performance targets were set arbitrarily low in order to facilitate achievement. Next, we consider how current performance compares to LOP targets, making a simple comparison of the achievement of the target thus far to the relative time elapsed under the activity. As Y2Q2 corresponds to 19 months of LINQ’s 39-month performance period (approximately 49% of total performance period expended), a useful shorthand may be to consider how current performance compares to LOP targets with reference to that figure; the same exercise is also conducted with reference to Y2Q3, corresponding to approximately 56% of total performance period expended. As shown in Table 14, performance with reference to this benchmark varies. • EQ2 >> Finding: LINQ IPTT data through Y2Q3 show that the activity is on track to meet less than half of its performance indicators, reflecting gaps in the effectiveness of the LINQ activity. o Performance approximately ‘on schedule’ – Indicators 4, 13 o Performance ‘ahead of schedule’ – Indicators 1 (non-capital), 2, 9, 10, 17 o Performance ‘behind schedule’ – Indicators 1 (total and capital), 3, 5, 7, 8, 11, 12, 13, 14, 15 40 The following analysis draws on information included in the LINQ Monitoring, Evaluation, and Learning Plan (February 2019 version) and MEL data reported within the Excel file “LINQ IPTT 2020 Y2Q2 Final” and the Word file “Y2Q3 LINQ IPTT table_Final.” 71 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT o Performance cannot be benchmarked (baseline is not available, results are not reported) – Indicators 6, 16 • EQ2 >> Finding: LINQ IPTT data through Y2Q3 show no results thus far for Indicators 7 and 8, measuring the effectiveness of USG in generating incremental sales at the firm level and at the farm level. • EQ2 >> Finding: LINQ IPTT data through Y2Q3 show no results thus far for Indicators 14 and 15, measuring the effectiveness of USG assistance in supporting access to finance by beneficiaries. • EQ2 >> Finding: LINQ IPTT data through Y2Q3 show that the LINQ activity has performed relatively well and ahead of schedule on those indicators that are outcome indicators, over which it has a greater degree of control (Indicators 9, 10, and 17). 72 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Table 16: LINQ Performance Indicators and Corresponding Disaggregation (Y2Q2 and Y2Q3) Indicator LOP Actual as % of LOP Target41 LOP Actual Disaggregation: Relative Shares per Indicator Y2Q2 Y2Q3 Y2Q2 Y2Q3 1. Value of new private sector investment leveraged by USG implementation - Investment Type: Capital - Investment Type: Non-Capital 46% 22% 71% 46% 22% 71% 23% 77% 23% 77% 2. Number of individuals benefitting from the activity - Sex: Male - Sex: Female 46% 74% 77% 23% 76% 24% 3. Number of people with improved incomes as a result of USG assistance - Sex: Male - Sex: Female 9% 12% 79% 21% 83% 17% 4. Proportion of female participants in USG-assisted programs designated to increase access to productive economic resources 24% 36% 5. Number of full-time equivalent jobs created as a result of USG assistance - Sex: Male - Sex: Female 13% 19% 27% 73% 41% 59% 6. Macroeconomic / financial situation developing in Lebanon conducive for investment N/A N/A 7. Value of incremental sales (domestic and export) collected at firm level for enterprises as a result of USG assistance - Size of MSME: Micro - Size of MSME: Small 0% 0% - - 8. Value of incremental sales (domestic and export) collected at the farm level for small holder producers as a result of USG assistance 0% 0% 9. Number of MSMEs, including farmers, and other organizations receiving business development services from USG - Size of MSME: Micro - Size of MSME: Small - Size of MSME: Medium - Size of MSME: Large 45% 79% 97% 2% 0% 0% 98% 1% 0% 0% 41 The benchmark for performance is 49% for Y2Q2 and 56% for Y2Q3. 73 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Indicator LOP Actual as % of LOP Target41 LOP Actual Disaggregation: Relative Shares per Indicator Y2Q2 Y2Q3 Y2Q2 Y2Q3 10. Number of individuals who have received USG-supported short-term agricultural sector productivity or food security training - Type of Trainer: F2F Volunteer - Type of Trainer: Local Volunteer - Type of Trainer: Consultant - Type of Trainer: LINQ Staff 54% 66% 6% 1% 79% 13% 6% 8% 68% 18% 11. Number of for-profit enterprises, producer organizations, water user’s associations, women groups, trade and business associations and community-based organizations (CBOs) that applied improved organization-level technologies or management practices because of USG assistance - Type of Enterprise: For Profit Private Enterprise - Type of Enterprise: Producer Organization - Type of Enterprise: Women’s Group - Type of Enterprise: Trade Business Association 34% 51% 67% 8% 25% - 50% 22% 19% 8% 12. Number of farmers or microenterprises who have applied new technologies or management practices as a result of USG assistance - Type of Entity: Farmer - Type of Entity: Cooperative - Type of Entity: Firm 30% 41% 66% 17% 17% 49% 20% 31% 13. Number of MSMEs, including farmers and other organizations, benefitting from new horizontal and vertical market linkages - Type of Linkage: Vertical - Type of Linkage: Horizontal 37% 54% 100% 0% 100% 0% 14. Number of enterprises that have successfully accessed loans, private equity, or both as a result of USAID 0% 0% 15. Total value of commercial loans accessed as a result of USG assistance - Size of MSME: Micro - Size of MSME: Small - Size of MSME: Medium 0% 0% - - - - - - 16. Subsidized loans accessed through banks and financial institutions N/A N/A 17. Number of learning and outreach events conducted, or materials developed by the activity 63% 91% 74 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT Of the 17 indicators, 11 are disaggregated according to relevant sub-populations on the basis of investment type, sex, business size, or type of trainer/enterprise/entity/linkage. LINQ performance to date for each of these 11 indicators is compared across the relevant disaggregation classes, and some distinct patterns are observed (Table 14). • EQ2 >> Finding: LINQ IPTT data through Y2Q3 show that LINQ has been more effective in leveraging non-capital investment by the private sector, than capital investment (Indicator 1). While LINQ’s performance for leveraging non-capital investment is ‘ahead of schedule,’ the capital investment leveraged is lagging behind. In absolute terms, LINQ had leveraged more than $1 million in non-capital investment as of Y2Q3, as compared to more than $300,000 in capital investment. This finding is further supported by the lack of results against Indicators 14 and 15, which indicate that no beneficiaries have been able to access alternative financing sources (commercial loan, private loan, or private equity). • EQ7 >> Finding: LINQ IPTT data through Y2Q3 show wide gaps in performance across gender. Men are overwhelmingly the primary beneficiaries of the activity (Indicator 2).42 Men also account for the clear majority of individuals enjoying increased incomes as a result of USG assistance (76%) (Indicator 3). Conversely, women account for a slightly larger share (59%) of FTE jobs created as a result of USG assistance (Indicator 5), albeit within a small number (9 FTE jobs). While women’s majority in this category initially appears surprising, it may possibly reflect widely reported wage discrimination in Lebanon that depresses women’s wages vis-à-vis those paid to men and perversely incentivizes their hiring over men. • EQ2 >> Finding: LINQ IPTT data through Y2Q3 show that LINQ has overwhelmingly targeted the micro and small categories of MSMEs (including farmers) through its provision of business development services (Indicator 9), which appears to reflect a commitment to support equitable, broad-based economic growth. • EQ2 >> Finding: LINQ IPTT data through Y2Q3 show that LINQ has heavily relied on consultants and, to a lesser extent, LINQ staff to deliver short-term training targeting agricultural sector productivity or food security (Indicator 10). The role played by F2F volunteers has been limited, which could plausibly be a reflection of travel restrictions linked to Lebanon’s revolution of October 2019 and the COVID-19 pandemic. • EQ3 >> Finding: LINQ IPTT data through Y2Q3 show that LINQ has not made extensive use of volunteer trainers to deliver trainings (Indicator 10), despite the fact that the use of volunteers could represent an opportunity to deliver training at lower cost and thereby contribute to greater efficiency of USG assistance. • EQ2 >> Finding: LINQ IPTT data through Y2Q3 show that for-profit private enterprises represent the largest share (50%) of those entities applying improved organization-level technologies or management practices as a result of LINQ assistance (Indicator 11), followed by cooperatives/producer organizations (22%), women’s groups (19%), and trade business associations (8%). 42 This figure is consistent with Indicator 4. 75 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT • EQ2 >> Finding: LINQ IPTT data through Y2Q3 show that farmers represent the largest share (49%) of those entities applying new technologies or management practices as a result of LINQ assistance, followed by firms (31%) and cooperatives (20%). • EQ2 >> Finding: LINQ IPTT data through Y2Q3 show that LINQ has been much more successful in establishing vertical linkages than horizontal linkages. IPTT indicators do not allow for a direct comparison between TA or training recipients, and the number of those that have applied improved organization-level technologies, new technologies, or management practices as a result of USG assistance, due to slight variations in the indicator definitions and the relevant population. To perform a more apt comparison, we therefore aggregate the number of TA or training recipients (across individuals and entities) and the number of individuals and entities applying the TA or training as shown in Table 15. Table 17: TA and Training Recipients versus TA and Training Application (Y2Q2 and Y2Q3) Indicator Number Proportion Y2Q2 Y2Q3 Y2Q2 Y2Q3 9. Number of MSMEs, including farmers, and other organizations receiving business development services from USG 1,263 2,221 10. Number of individuals who have received USG￾supported short-term agricultural sector productivity or food security training 1,188 1,446 Sub-Total 2,451 3,667 11. Number of for-profit enterprises, producer organizations, water user’s associations, women groups, trade and business associations and community-based organizations (CBOs) that applied improved organization-level technologies or management practices because of USG assistance 24 36 12. Number of farmers or microenterprises who have applied new technologies or management practices as a result of USG assistance 320 427 Sub-Total 344 463 TOTAL 14% 13% • EQ2 >> Finding: LINQ IPTT data through Y2Q3 show that USG assistance has been marginally effective in delivering TA or training assistance that recipients have applied to date. • EQ2 >> Finding: There is a wide gap between self-reported data through the farmer/trainees survey and the formal MEL reporting indicators on the application of training delivered with USG assistance. 76 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 12. SUPPLEMENTAL QUESTION: TRADE FINANCE When was the last time your business accessed or obtained trade finance (e.g., letter of credit)? Where did you source this trade finance? What are your/the agri-food sector’s challenges in accessing trade finance currently? FINDINGS 1. Those beneficiaries who offered a substantive response on the issue of access to trade finance generally agreed that access to trade finance had become more problematic for the agri-food sector over the past year. “It is not only hard [to access trade finance in the past 12 months], it is impossible” (Beneficiary-Grantee, Food Processing Sector). 2. To cope with the lack of access to trade finance, businesses have reportedly adapted in several ways. These include: considering reducing staff to cut expenses and reserve available funds; reducing business lines that depend on access to hard currency; relying on offshore accounts (theirs or of business partners, family, friends); relying on export earnings to deposit funds in an offshore account or to earn “fresh money”; and converting to cash-based sales. One beneficiary had adopted alternative arrangements to manage this issue but was uncomfortable discussing it over the phone with the evaluation team. 3. For those businesses that could potentially benefit from an effort by the government (the Ministry of Economy and Trade) to facilitate access to hard currency for imported inputs, it remains unclear whether they will in fact benefit as intended. Beneficiaries who spoke about this expressed skepticism that it would benefit agribusinesses. 4. Several beneficiaries suggested that USAID assist in securing access to hard currency to facilitate international trade, though they offered no specific ideas on how this could be provided. 5. One respondent spoke of diaspora finance as a source of investment in Lebanon’s agri-food sector. CONCLUSIONS Access to trade finance is indeed a problem for beneficiaries, one that has intensified in the past 12 months. Beneficiaries are working to find creative solutions through the mobilization of diverse financial resources to overcome this issue. However, the beneficiaries are incurring some considerable transaction costs to use these creative solutions. They have little confidence in the solutions proposed by the government to deliver hard currency to agro-processors and which have not yet been implemented. Conversely, possible USAID assistance in this area is viewed positively by grantees and non-grantees. RECOMMENDATIONS 1. USAID should conduct further analysis to develop concrete options to support access to trade finance by beneficiaries within the agro-processing sector, possibly including through letter of credit and/or letter of guarantee and/or diaspora financing. 77 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 13. POST-BEIRUT BLAST ECONOMIC AND OPERATIONAL OVERVIEW The Port of Beirut suffered extensive damage as a result of a massive explosion that occurred on August 4, 2020. The immediate implications for the sea-based import and export of all goods (including agri￾food products) is unclear, and have raised questions concerning the findings, conclusions, and recommendations made elsewhere within this LINQ Mid-Term Performance Evaluation. The following information was obtained through a rapid assessment targeting key informants in trade and specifically trade of agri-food products, to understand the current status and implications of sea￾based trade after August 4, 2020. The rapid assessment was conducted by Bashar Berro by telephone/WhatsApp between August 7 and August 10, 2020. [name redacted] • The Port of Tripoli is ready and equipped as informed. • Tripoli port is capable of hosting big vessels to load containers. • Imported cows and livestock vessels still not informed where to go. • As a syndicate, we have members that have imported containers and containers for export at Beirut port and still have no exact idea about their condition. • We are not worried much about trade cycle, expecting to have everything normally running soon. [name redacted] • Beirut port is currently operating. • CMA CGM to start back operating on August 12 and MAERSK on August 15. • Bay 5 is ready to host big vessels. • Private companies like CMA CGM got their own electric generators to supply their refrigerated containers and vessels. • Tripoli port cannot host big vessels with its 12 meters width of parking side. • Beirut port can host big vessels with its 24 meters width of parking side. • Clearing agents are back to their offices in the Port of Beirut. • MAERSK started disembarking containers from port to customers in Lebanon in order to load. [name redacted] • Beirut port will gradually get back operating starting August 10. • Bays 12, 13, and 14 will be dedicated to bulk carriers, roll on-roll-off (RO-RO) carriers, and livestock carriers. • As big cranes are destroyed at some bays, clearing agents are requested to as gearless vessels to arrange/rent private cranes to load and offload upon arrival to Beirut Port. • Temporarily, entrance to Beirut loading bays is from Gate 14 Karantina. [name redacted] • Nothing is clearly visible yet. • Beirut Port is almost destroyed from Bay 12 to 19. • Gate 3 to Port Administration is almost ruined. • Administration to Bay 14 area seems safe but needs inspection for ground and cranes. • Many offices are destroyed. 78 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT • Partial shipments are difficult now to get offloaded and stored. • Expected in a month to have new warehouses. • Full imported containers need fast inspection. • Bureaucratic clearance is not useful nowadays, need to rely on documents and analysis/certificates from importing destinations with good reputation to clear fast. • Top managers at Beirut port are arrested now, and temporary managers are hired. [name redacted] • 160 fresh containers were parked at Beirut port during explosion. • The Syndicate made an effort to connect the reefers (cooled containers) with electricity with available resources. • Communicated with Minister of Transportation to assure that Customs, General Security, and Ministry of Agriculture have their offices back to stamp and facilitate export procedures and documentation. • This week a vessel will load and embark to Jeddah, Saudi Arabia and Jebel Ali, Emirates. • Another vessel this week will embark to Egypt. • Vessels aiming to load from Beirut started parking at Beirut Port Bays as of August 10. [name redacted] • Beirut Port is in operation again. • Exporters are in process of preparing certificates of origin again aiming to keep exporting and having fresh USD in the country. • 60% of Beirut Port is operating as a general estimate. • The main index of operations volume is the certificates of origin processed at the four chambers of commerce in Lebanon (Zahle, Tripoli, Beirut, and Saida). [name redacted] • Gradually CMA CGM willing to be back. • Currently working from home. • We are operating from Tripoli Port. • Not all vessels are able to park at Tripoli Port. • No official clearance from Beirut Port to us as CMA CGM to operate again into Beirut Port. [name redacted] • Most of cranes at the Port of Beirut are operating. • There will be some stress in operations as some bays are destroyed. • Customs are now operating from Beirut Airport and corresponding to documents and shipments in Beirut Port. • Lebanese Army is still restricting some employees and people to arrange some bays. • Clearing Agents are expecting to operate back again by the third week of August. • Tripoli Port is operating. • We are focusing more on land border (Al Masna’a) as an export channel. 79 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 14. BIBLIOGRAPHY EnCompass, LLC & Social Impact, Inc. (2019, January 16). USAID Lebanon Gender Assessment Report: PMSPL II. Land O’Lakes International Development. (2019, February 20). Lebanon Investment in Quality (LINQ) Monitoring, Evaluation, and Learning Plan. Land O’Lakes International Development. (2018, December 20). Lebanon Investment in Quality (LINQ) Sector Assessment. Land O’Lakes International Development. (2018, November 16). Lebanon Investment in Quality - LINQ Program: Gender Integration Strategy. Land O’Lakes International Development. (n.d.). LINQ Proposed Action Plan to Address the COVID-19 Outbreak in Lebanon. Land O’Lakes Venture 37. (2019, October). Lebanon Investment in Quality (LINQ) Cooperative Agreement No. 72026818LA00001: Annual Progress Report, October 1st 2018-September 30th 2019. Reuters. (2020, August 13). “Hardened Lebanese merchants battered by Beirut blast.” The Daily Star. http://www.dailystar.com.lb/Business/Local/2020/Aug-13/510259-hardened-lebanese-merchants￾battered-by-beirut-blast.ashx Social Impact, Inc., Bahn, R., Bayram, G., & Abou Jaoude, H. (2019, February). USAID Lebanon Economic Growth Assessment: Final Assessment Report. Available at: https://pdf.usaid.gov/pdf_docs/PA00TQ5W.pdf. 80 | LINQ MID-TERM EVALUATION USAID/LEBANON SEPTEMBER 2020 - FINAL REPORT ANNEX 15. DECLARATION ON CONFLICT OF INTEREST The authors of this report declare that they have no conflict of interest. Social Impact, Inc. 2300 Clarendon Boulevard Arlington, VA 22201 Suite 1000 Tel: (703) 465-1884 Fax: (703) 465-1888 www.socialimpact.com