Our Case Studies

Using mobile phones to track ongoing trends in food market functionality
A case study with FEWS NET in Sierra Leone and Liberia

Background

FEWS NET, the Famine Early Warning Systems Network, is a leading provider of early warning and analysis on food insecurity. Created by USAID in 1985 to help decision-makers anticipate humanitarian crises, FEWS NET provides objective, evidence-based reporting on more than 36 of the world’s most food insecure countries.

As the Ebola outbreak of 2014 continued to spread, FEWS NET had a need to ram up their ongoing monitoring of food security in the affected countries of Sierra Leone and Liberia. Population movement, quarantine, and other factors had diminished food availability and disrupted markets. However, due to the contagious nature of Ebola, FEWS NET was unable to conduct in-person market assessments in these countries. With an urgent need for data, FEWS NET turned to GeoPoll to conduct remote SMS surveys in Sierra Leone and Liberia. FEWS NET analysts hoped to create a panel of market traders in each country who they could survey on an ongoing basis about market sizes, activities, operating costs, stock levels, and agricultural activates.

The SMS-based survey results serve to corroborate key partner reports on market activities, serve as inputs to FEWS NET’s food security analysis on the impacts of the Ebola outbreak, and are used for emergency planning in response to the outbreak.

Solution

From November 2014 to October 2015, GeoPoll and FEWS NET conducted bi-weekly surveys in all 14 districts of Sierra Leone and 15 counties in Liberia.

In the first round of data collection, GeoPoll was tasked with building a panel of traders or merchants who trade palm oil, cassava, imported rice, or local rice. To do this, GeoPoll sent survey invitations to a random sample of its database of over 1,615,000 in Liberia and over 1,844,000 in Sierra Leone. This sample was asked if they are a trader or merchant of one of the previously mentioned commodities. Those determined to be eligible were added to the panel and asked initial questions about the typical amount of their trading, weekly inventory compared to others in their area, and where their customers and suppliers were located at time.

In the following rounds of data collection, the panel was sent bi-weekly surveys asking about the opening status of markets in their area, food availability, quantities of palm oil, local and imported rice, and cassava in markets, and where most suppliers and customers were located in the past 10 days. Questions were also asked about opportunities to earn wages on farms, crop cultivation methods, and cocoa, rubber, and coffee sales in their area. Each round of data collection included responses from between 300-500 market traders, allowing FEWS NET to view longitudinal data from the same group of respondents.

Results

Findings demonstrate that the Ebola outbreak impacted food availability, market operations, and agricultural activities. Throughout data collection, the situation grew starker in Sierra Leone, where there was for some time an official ban on markets. At the beginning of data collection, market closures in Sierra Leone were reported at 10%, which dropped to only 5% by round ten. However, regularly 40-60% of traders reported markets operating at reduced capacities. In later rounds of data collection, around 60% reported reduced opportunities to earn wages, and just 30-40% reported that agricultural activities were proceeding on time.

In Liberia, only a very few traders indicated market closures (less than 1%), and between 20-30% reported reduced capacities at markets. However, in the most Ebola-affected areas between 25% and 40% reported that food availability did not meet local needs. 45-50% reported agricultural activities were going forward as normal and on time.

Below are findings from several rounds of data collection. Data and reports can be viewed in more detail on the FEWS NET pages for Sierra Leone and Liberia.

Key Points

  • Bi-weekly remote mobile surveys track the functionality of food markets in Sierra Leone and Liberia.
  • Each round of data collection included responses from between 300-500 market traders, allowing FEWS NET to view longitudinal data from the same group of respondents.
  • Data collected includes information on market operating status, stock levels of specific good, and agricultural activities.
  • Survey results corroborate partner reports on market activities and serve as inputs into FEWS NET's food security analysis on the impacts of the Ebola outbreak.

Be Notified Of New Case Studies

Download PDF Report

Related Links

Sierra Leone Findings

Media Measurement

The second round of data collection was conducted November 3-17, 2014, 2014, and included a sample of 400 traders from the initial panel:

  1. Despite an official ban on weekly markets, over 90% of traders reported that markets were open in their areas.

  2. Over 40% reported that markets were operating at reduced capacity.

  3. In 6 districts (Bo, Bonthe, Kambia, Kenema, Moyamba, and Port Loko) 50% reported below average stock levels for common goods.

The fourth round of data collection was conducted the week of December 22, 2014 with a sample of over 300.

  1. Similar to the second round, only 10% reported weekly market closures.

  2. Nearly 60% reported markets were operating at reduced levels.

  3. Despite being in a post-harvest period, nearly 40% said market supplies of main commodities had declined since early December.

  4. 50% of traders reported food supplies did not meet local needs.

  5. Travel restrictions were the most commonly cited reason for reduced market supplies.

Later rounds of data collection asked about both market activities and agricultural activities. The eighth round was conducted during the week of March 30th, with a sample of 500 traders.

  1. In late March, 10% still reported market closures, and 53% reported markets operating at reduced levels.

  2. 40% indicated lower levels of market supplies, and travel restrictions continued to be the main reason for this.

  3. Over 75% indicated that the current primary agricultural activity is land preparation, and just 32% reported normal and on-time agricultural operations.

  4. Over 64% reported reduced opportunities for wages, and over 35% reported not being able to sell cash crops as usual.

In the tenth round of data collection, conducted the week of May 25, 2015, with over 400 traders, market closures were diminishing:

  1. 5% reported market closures and 41% reported reduced levels of operation.

  2. 35% indicated lower levels of market supplies, and high transport costs were the most cited reason.

  3. 51% indicated that the primary agricultural activity was land preparation, and 39% reported normal and on-time agricultural activities.

  4. More than 58% reported reduced wage opportunities and 37% reported not being able to sell cash crops.

Liberia Findings

Media Measurement

The third round of data collection in Liberia collected data the week of December 8, 2014, from over 500 traders.

  1. Over 30% reported that weekly markets operated at reduced levels, however only 7 (less than 1%) reported weekly market closures.

  2. Despite the post-harvest period, 25% indicated that supplies of main commodities had declined since November.

  3. In areas hardest hit by Ebola, between 25% and 40% reported that food availability at markets was insufficient to meet local needs.

  4. High transport costs was the most cited reason for reduced supplies.

Round five of data collection collected data from over 300 traders in the week of January 5, 2015.

  1. During the first week of January, 23% reported reduced levels of operation at weekly markets, while only 2 traders reported market closures.

  2. 25% of traders indicated the market supplies of main commodities had declined since late December.

  3. 25% - 41% in areas most affected by Ebola reported food availability did not meet local needs.

  4. High transport costs continue to be most frequently cited reason for reduced market supplies.

Later rounds of data collection also asked about agricultural activities. In round eight of data collection, GeoPoll reached 500 traders across all 15 counties in Liberia:

  1. 26% reported markets in their area operated at reduced levels, and only 4 traders reported market closures.

  2. Over 30% reported reduced availability of main market commodities, with high transport costs the most cited reason for the reduced supplies.

  3. 48% indicated the primary agricultural activity is land preparation, and 46% reported normal and on time agricultural activities.

  4. 39% reported less wage opportunities than normal, and over 35% reported not being able to sell cash crops.

In round ten, data was collected during the week of May 25th with a sample of 435 traders.

  1. At the end of May, 21% reported reduced market activities and only one trader reported market closures.

  2. Over 23% reported supplies of main commodities were diminished, with high transport costs continuing to be the most cited reason for shortages.

  3. More than 44% said planting was the current primary agricultural activity, and 49% reported normal and on-time agricultural activities.

  4. 33% reported less wage opportunities than usual, and over 17% reported not being able to sell cash crops as usual.

  • Top
  • 3000 Lawrence Street
    Ste 125
    Denver, CO 80205
  • Washington, DC
    2000 M St. NW #790
    Washington, DC 20036
  • Nairobi, Kenya
    Mobile Accord
    The Mirage, M2, Suite 11
    Chiromo Rd, Westlands