As COVID-19 spread to many low and lower-middle income countries in March and April and control measures were imposed, researchers began monitoring impacts on income and food security. Many relied on phone surveys to capture the initial effects. As time passes, follow-up surveys can help to refine those initial observations and account for changing circumstances. In this post, Alan de Brauw, Kalle Hirvonen and Gashaw Abate report on income and food consumption patterns in 600 households in Addis Ababa, surveyed monthly from May to August—updating a May post on their initial findings. Under modest movement restrictions, households have stabilized incomes and overall food consumption relative to initial, more pessimistic perceptions. The authors show the benefits of continuing to survey affected populations as the pandemic evolves, and the need for additional income and price data to supplement phone surveys to better inform recovery actions.—John McDermott, series co-editor and Director, CGIAR Research Program on Agriculture for Nutrition and Health (A4NH).
There is substantial concern that global food insecurity is increasing as a result of the COVID-19 pandemic. Organizations such as the UN Food and Agriculture Organization (FAO), World Food Programme (WFP), and CGIAR are trying to get a clear picture of the growing challenges people may be facing, but the nature of the pandemic makes data on food security and daily living challenging to obtain. Because face-to-face surveys have largely not been possible during the pandemic, much of what we know about actual changes in food security status is coming from phone surveys. These surveys suggest large proportions of respondents are facing falling incomes; the World Bank reports declines in incomes and food security where they have conducted phone surveys.
These reported income changes suggest the pandemic has affected a large proportion of the world’s population and that the poor in particular face the threat of significant food insecurity. However, phone surveys typically do not allow us to quantify respondents’ changes in income. In other words, we know they perceive their incomes decline, but we do not know how much they have actually declined. As a result, analysts must make assumptions about the severity of income changes.
A series of surveys we have been conducting in Addis Ababa, Ethiopia suggests there may be reason to be cautious about such assumptions. While a majority of people have reported lower incomes in recent months, overall food consumption in our most recent survey (conducted in early August) was comparable to pre-pandemic levels, though with shifts in dietary composition. These results, detailed in our new discussion paper, indicate that it is easy to read too much into survey questions regarding subjective changes in incomes in the pandemic, and also that many food value chains in Ethiopia have continued to function relatively well in the crisis.
To understand how information on income changes corresponds to changes in food consumption, we have tracked a panel of households in Addis Ababa since 2019. Panel households were initially surveyed in person about their food consumption in Aug. and Sept. 2019, and were tracked through the COVID-19 pandemic through phone surveys in May, June, and July 2020. The Aug. 2020 food consumption survey allows us to directly compare consumption during the pandemic with consumption at the same time of year in 2019, six months before the pandemic began in Ethiopia.
Collectively, the surveys suggest that even if incomes have declined among a majority of the city population, the value of food consumption has not. The Aug. survey this year found that overall food consumption was similar to pre-pandemic levels. The distributions of values are nearly on top of one another, suggesting the food budget is about the same among households across income groups—even among poorer households. Moreover, we do not find a decline in food consumption among households that had reported an income decline or a lost job in the previous (July) phone survey, implying that the subjective question about an income loss does not predict changes in food security levels.
Although this finding is good news, we do find a shift in the distribution of expenditures between categories of foods (Tables 1 and 2). Specifically, grain and staple consumption has risen on average, in both value and calorie terms, while vegetable consumption and calories consumed among “all other foods,” including oils, have fallen. Although the decline in consumption of vegetables, which are high in micronutrients, is potentially concerning, consumption of other micronutrient-dense foods, such as animal sourced foods and fruits, has remained steady on average.
Taken together, these results have two important implications. The first result casts some doubt on the usefulness of the types of questions being asked about subjective changes in incomes, at least in studying changes in household well-being during the pandemic. Note that even in our July survey, 64% of households reported an income decline relative to normal levels at that time of year. However, the questions we and others typically ask do not allow us to quantify the income loss. Our results suggest that simply reporting the proportion of households that say their incomes have declined is misleading in terms of welfare and poverty impacts, and may lead some to seriously over-exaggerate the welfare and poverty impacts of the ongoing pandemic.
The shift in consumption patterns within the overall budget, meanwhile, suggests that food prices and/or the availability of certain foods may have played an important role in changing consumption patterns. As overall consumption was mostly unchanged year-over-year, it is clear that food is available and many food value chains have either been or become quite resilient to the economic shock associated with the COVID-19 pandemic in Ethiopia.
For these value chains, the pandemic may have caused a temporary shock—given initial disruptions in supplies, people had to figure out new ways of interacting to exchange foods and food products as food moved from producers to consumers. In the initial phases of the crisis, our May survey showed declines in fruit, meat, and dairy consumption. We surmise that there were initially issues with the supply chain for those foods. Once those issues were resolved, overall consumption could proceed largely as it had before the pandemic. It is also worth mentioning that Ethiopia never imposed a full lockdown that severely restricted movements, which may have played a role in the resilience of some value chains or their ability to bounce back relatively quickly. Experiences might differ in countries that imposed more complete lockdowns.
However, the decline in consumption of some food categories suggests that not all value chains have been equally resilient. To the extent that these foods may play an important role in micronutrient consumption, understanding what is happening with those value chains and how to help them recover is crucial. A first step would be to analyze price data as they become available. Learning where prices have risen fastest during the crisis might help us better understand factors that might lead value some chains to break during a crisis while others continue to function.
Alan de Brauw is a Senior Research Fellow with IFPRI’s Markets, Trade and Institutions Division (MTID); Kalle Hirvonen is a Senior Research Fellow with IFPRI’s Ethiopia Strategy Support Program (ESSP); Gashaw Tadesse Abate is an MTID Research Fellow. The analysis and opinions expressed in this piece are solely those of the authors. This work was funded by the IFPRI-led CGIAR Research Program on Agriculture for Nutrition and Health (A4NH).
