Finance & PSD Impact AUGUST 2019 The Lessons from DECFP Impact Evaluations ISSUE 54 Our latest note examines the impact of a social enterprise’s attempts to shorten the supply chain between farmers and small retail stores in Bogota, an idea that was one of the winners of the SME ideas competition. Shortening Supply Chains for Fruit and Vegetable Vendors in Bogota Leonardo Iacovone and David McKenzie The small retail stores that are prevalent this idea. They agreed to an impact evaluation on many street corners in developing to measure the impacts of this new service. countries face very different supply chains for different products. On one hand, Experimental Design multinational suppliers of branded non- Agruppa went door-to-door in poor perishable products have amazingly efficient neighborhoods of Bogota in Jan-Feb 2016 to supply chains that result in store-door map stores selling fruit and vegetables. Using delivery to even the most remote places. In larger streets as natural boundaries, these contrast, the fresh fruit and vegetables they neighborhoods were divided into 63 market sell are part of what are typically much less blocks, containing 1,620 firms. These blocks efficient supply chains, that often end in were then randomized into 32 treatment centralized wholesale markets, requiring blocks and 31 control blocks. substantial travel time and costs for retail Baseline surveys in these blocks collected store vendors, and increased food costs for information on the firms and their owners, the urban poor. and at the end, firms received a factual This is the case in Bogota, where the explanation of Agruppa, and were asked typical store owner gets up at 4:30am each whether they would be interested in being a morning, and spends an average of 2 hours client should Agruppa launch in their block. travelling to, around, and back from the Our main analysis compares the 586 central market of Corabastos in order to buy interested firms in treated blocks who get fruits and vegetables. offered Agruppa to 536 interested firms in control blocks who are not offered Agruppa. A social enterprise proposes a solution We measure usage of Agruppa using The social enterprise Agruppa was administrative data on purchases, and formed with the goals of using new conducted seven rounds of high-frequency technologies to overcome this problem, follow-ups at 2, 4, 6, 10, 14, 26 and 52 weeks reducing travel time and purchase prices for after Agruppa had launched in a block in store owners, and in turn reducing the prices order to measure impacts on stores. charged to poor households who rely on these stores for their fresh foods. Take-up and Usage Agruppa’s idea was to use technology to Agruppa launched by selling only five of the agglomerate orders each day from many most frequently purchased products focusing small retailers, source in bulk directly from initially on the bulkier and heavier rotation farmers where possible, and then deliver the ones (potatoes, plantains, tomatoes, onions, produce to the store fronts of retailers. and spring onions). These core products The founders pitched this idea at a 2015 accounted for just over half of total sales in SME ideas competition held by the World the retail stores at baseline. They gradually Bank, and received some seed funding that, increased their products over time, reaching coupled with funding they raised from social 28 products during our survey period. investors, enabled them to attempt to scale Do you have a project you want evaluated? DECRG-FP researchers are always looking for opportunities to work with colleagues in the Bank and IFC. If you would like to ask our experts for advice or to collaborate on an evaluation, contact us care of the Impact editor, David McKenzie (dmckenzie@worldbank.org) Figure 1 shows take-up and usage of • Retailers save on purchase costs by 6 to Agruppa in the treatment blocks. Initial 8 percent, they pass some of this onto interest is high, with 52% of interested firms consumers in the form of slightly lower making at least one purchase from Agruppa prices, but also increase mark-ups. within the first two weeks, and 66% making • However, overall sales and profits a purchase at least once in the first year. But appear to have fallen, due to less sales after this initial interest, usage tapers off over of non-Agruppa products. This finding time, so that only 24% are still using Agruppa should be caveated by many firms not after 6 months and only 16% after one year. revealing their sales. It is consistent with retail firms being less likely to be Figure 1: Declining usage over time carrying some of the fruits and vegetables that were not core-Agruppa products. • Our results show firms in this sector are not in perfect competition, with wide cross-sectional price dispersion. This limits the gain in sales to be had from small reductions in prices. Policy Implications 1. Agruppa was not able to scale-up its model fast enough to break-even, and closed as a business on January 20, 2018. Impacts on Retailers 2. Nevertheless, the underlying idea of • Agruppa’s service did reduce travel shortening value chains in this way shows time and costs, just not as much as promise, and at least two new enterprises, anticipated. Over the first six weeks of introducing Agruppa, trips to Corabastos Frubana in Colombia, and Twiga in market fell 0.4 days per week for those Kenya, are pursuing similar ideas. using Agruppa, and conditional on going 3. This case also offers lessons for start-ups to market, firm owners spent less time and researchers considering there. As a result, those using the service implementing randomized trials during saved almost 2 hours per week in travel the early stages of business launch. There time, relative to a control mean of 12 were several advantages of collaborating hours. Accordingly, firm owners continued to travel frequently to market at this early stage, but also challenges. to buy other products not sold by Future collaborations may benefit from Agruppa. waiting until proof-of-scale can be • Work-life stress improved, particularly delivered in one location before then measures of having time for family and evaluating expansion to another location. for all the activities they have planned. For further reading see: Iacovone, Leonardo, and David McKenzie. 2019. “Shortening Supply Chains: Experimental Evidence from Fruit and Vegetable Vendors in Bogota”, World Bank Policy Research Working Paper no. 8977. Recent impact notes are available on our website: http://www.worldbank.org/en/research/brief/finance-and-private-sector-impact-evaluation- policy-notes