Background
Agricultural value chains (AVCs) are a key component of the food systems and the broader economies of countries around the world. While the needs of farmers and consumers have been studied from many perspectives, the financial activities and needs of โmidstreamโ firms, which move agricultural products from the farm to retail markets, are poorly understood. Yet these actors play an important role in providing reliable consumer access to food (particularly in urbanizing areas), creating non-farm employment opportunities, and generating export revenue.
In an ongoing project funded by the Gates Foundation, we are building an understanding of the financial strategies and unmet needs of midstream firms. In previous work we found, among other results, that intermediary firms overwhelmingly use cash to facilitate their business transactions, and though we do observe that many intermediary actors are familiar with electronic bank transfers or mobile money, these digital payment systems are used very infrequently (see Food Policy paper here).
Scope of work
The current project aims to develop an understanding of the relevant constraints on the use of digital payment systems among producers and intermediary firms within agri-food value chains. We are planning to implement focused surveys, using the sampling methodology developed in the previous project (see IFPRI Discussion paper here), among intermediary agri-food value chain actors in the poultry and maize value chains in Tanzania. These focused surveys will more specifically document:
The use of digital financial services and digital payment systems
The risks and/or limitations of cash-based payments
The constraints on the use of digital financial services and digital payment systems disaggregated by role within the value chain.
This terms of reference seeks a survey firm to conduct the data collection in Tanzania. Survey work will follow the poultry and maize value chains with focus on two to three regions of the country.
The goal of this sampling approach is to systematically survey producers and the โmidstreamโ of specific agricultural value chains so to understand the extent of financial access among producers, agricultural traders, transporters, processors, aggregators, and wholesalers. This is a challenging task because agricultural value chains take the form of a network whereby actors at various stages of the value chain transact with each other in complex ways. Thus, we will implement the network-driven sampling approach, as described in this IFPRI Discussion Paper and as summarized below.
Interview Process and Logistics
The interview process follows an iterative approach and aims to capture information about actors within multiple segments of the midstream agricultural value chain for a specific commodity. The interview process aims to identify producers and midstream actors at various segments of the value chain, each representing key levels of value addition. These segments include (but is not necessarily limited to): (i) farmers, (ii) traders, (iii) processors, and (iv) wholesalers.
It is important to note that in some cases these segments may be combined into one firm (i.e., a firm that both processes and sells wholesale) and in other cases there may be additional links to other firms (i.e., transporters). Our goal is to develop an interview protocol that is flexible enough to incorporate these details while also sufficiently standardized so to collect comparable information across value chains and country contexts.
The iterative data collection process starts with selecting a small number of farmers in randomly selected villages within each of the districts selected for data collection. We first interview farmers, and the survey includes questions asking them to whom they sold the commodity of interest in the last 30 days, and record the name and phone number of these selling links. These reported names and phone numbers form a database of referred intermediary actors. Next, we contact the midstream actors identified by the farmers and interview them. We again record the name and phone number of these selling links, add these referrals to the intermediary actor database, and interview the identified actors if they have not yet been interviewed. This iterative process continues until we either achieve โsaturation,โ whereby referred intermediary value chain actors are already surveyed and included in our data, or we are referred to intermediary value chain actors that are outside the scope of our survey either geographically or because we reached the retail segment of the value chain. The survey instruments include information on buying and selling commodities, costs, financial access, credit, payments, contracts, and others. Interview length is expected to be approximately between 1 and 1.5 hours on average.
Logistics
The network-driven sampling approach requires an iterative sampling strategy. Note that while we ask about backward links, to simplify the logistics of our interview plan, our sampling strategy will only follow forward links and progress through the value chain starting from production and moving toward consumers. The logistics of this approach are as follows:
Step 1: Interview farmers and record where they sell the commodity of interest. With these responses by farmers, we construct a roster of midstream actors.
Step 2: Begin interviewing midstream actors reported in step 1. Record where they sell the commodity of interest (i.e., local markets, cooperative organizations, other traders, processors, wholesalers), and contact information for those sales. It is important to set up a database of those contacts that is updated in real-time (or near real-time), so that the database can be checked for duplication before setting new interviews. It is helpful to keep duplicates in the database.
Step 3: Continue interviewing midstream actors reported in steps 1 and 2. Record where they sell the commodity of interest. Continue interviewing midstream actors until we reach saturation (i.e., there are no more un-surveyed midstream actors within geographic scope who will talk to us), or we hit a target number/budget constraint.
Please take careful note of these logistics. This survey is non-standard and distinct from typical surveys that have a full list of sampled households/individuals prior to the start of enumeration. This sampling approach requires that the roster of sampled midstream actors is dynamic and generated in real time while enumeration is ongoing. Based on our experience, it is useful for the survey firm to allocate at least one individual who manages the midstream actor roster in real-time and coordinates with enumerator teams to facilitate efficient fieldwork.
Target Survey Numbers and Budget Constraint
The budget ceiling for this survey is $155,000 USD.
Using the above approach, we expect to complete 2,000-3,000 total interviews (across all types), beginning with about 250 farmers per value chain.
Survey Instrument
The survey will consist of the following topics and take approximately between 1 and 1.5 hours to complete.
Access to and use of formal financial services
Access to and use of informal financial arrangements
Access to and use of mobile money to facilitate financial transactions
Demographic information about the respondent
Forward and backward links in the coffee value chain
Scale of business
Amount (i.e., kgs) of coffee flowing through their organization
Number of employees (if any) and wages paid to employees (if applicable)
Enterprise revenue, costs, and profit
Barriers on the adoption and use of digital financial services
Survey Firm Tasks
Obtain necessary ethical approvals and research clearances
Comment on the draft survey instruments provided by the research team.
Translate and back translate the survey instruments.
Program the survey instruments into Survey CTO, test the survey instrument, make the programs available to the research team for testing, and respond to comments from the research team.
Conduct pilot of survey instruments
Create survey training plan and manual.
Conduct survey training for enumerators and other staff.
Develop survey logistics plan with input from the research team.
Develop quality control plan, subject to research team approval.
Carry out all quality control measures.
Provide daily updates to the research team during survey work.
In support of these activities, the IFPRI research team will:
Provide the survey firm with the regions of field work for each value.
Collaborate with the survey firm to develop the sampling frame of villages and respondents.
Collaborate with survey firm to finalize the sampling strategy.
Provide the survey firm with draft survey instruments and engage with the survey firm during piloting to update them as needed.
Deliverables
Translated survey instruments
Draft Survey CTO instruments
Final Survey CTO instruments
Survey training plan
Survey logistics plan
Survey quality control plan
Raw datasets, including database of forward links between actors
Clean datasets with labeled variables
Regular updates
Required qualifications
Ability and appropriate authorizations to carry out data collection work in Tanzania
Experience collecting in-person survey data in Tanzania, ideally in rural areas
Project managers should be fluent in English and able to communicate verbally and in writing with the research team
To apply: Please apply with a short proposal that includes a proposed data collection plan, estimated timeline, a proposed budget, details on the firm and qualifications/CVs of primary team members. Direct any questions to: Kate Ambler (k.ambler@cgiar.org) and Jeff Bloem (j.r.bloem@cgiar.org). Deadline for submission of proposals: November 2.