Why Pandemics Don’t Always Increase Food Insecurity

Basic economic theory says that public health lockdowns, such as the ones being imposed during the COVID-19 pandemic, are likely to raise food prices and aggravate food insecurities in vulnerable communities. The resulting decline in market activity and the interruption of supply chains are expected to decrease household income — especially in informal labor markets — and increase food prices due to transit restrictions and costs arising from new quality controls and guidelines. The stress is particularly pronounced for low-income households in developing countries, where lockdowns and activity restrictions are often unaccompanied by safety net policies designed to mitigate their negative impacts. But in a recent working paper, Aggarwal et al. find no evidence that market disruptions worsened food security in rural areas.

To study the effects of market disruptions on food security, Aggarwal et al. turned to an unconditional cash transfer program which was instituted prior to the pandemic. The UCT targeted rural, low-income households in Liberia and Malawi. Though only Liberia had instated a lockdown, markets were disrupted in both countries following the pandemic, suggesting that declines in economic activity can to a large extent be explained by voluntary adjustments to behavior in response to the pandemic. The researchers analyzed household survey data collected during the UCT program and found no negative effect on food prices and food consumption as a result of the implementation of restrictions from March to July of 2020.

Although food prices did change over the course of the study, such changes appeared to be seasonal rather than a consequence of the pandemic: when seasonality is controlled for, food prices trend negative in both Liberia and Malawi over the April 2020 to August 2020 period, relative to February prices. During the pandemic, food vendors had difficulty maintaining sufficient stock levels and suffered monthly profit declines of 42% in Malawi and 52% in Liberia. However, according to the researchers’ survey, rural household access to food was not affected by this phenomenon. One possible explanation for this is the fact that rural areas have lower population densities: consequently, program participants are less exposed to the virus, and are likewise cushioned from the effects of economic disruption.

Finally, the researchers found that additional cash transfers did not significantly improve food security. Aggarwal et al. observed that the cash transfers resulted in increased food consumption and improved food intake quality; crucially, however, food security did not worsen for the control group. From this, the researchers concluded that lockdowns did not increase food insecurity among the rural households surveyed. Since food insecurity was a prevailing problem in the UCT program areas prior to the pandemic, the extent to which UCTs can mitigate it is limited. For instance, nearly 25% of participants reported having no food for an entire day during the months in which the survey was conducted.

This research is an important step forward in our understanding of the efficacy of safety net policies in the face of crisis-induced market disruptions. Although there was no evidence that UCTs directly mitigated food insecurity, the increase in food intake and quality for program participants shows that better safety net policies can make a difference. Since this finding is specific to rural areas, future research should study the effect of UCT programs on food security among urban populations. Additionally, in contrast to the relatively short timeline of this study, it is vital that subsequent studies track the long-run impacts of UCTs, since many of the consequences of the pandemic have yet to be observed.

---

Aggarwal, Shilpa, Dahyeon Jeong, Naresh Kumar, David Sungho Park, Jonathan Robinson, and Alan Spearot. 2020. “Did COVID-19 Market Disruptions Disrupt Food Security? Evidence from Households in Rural Liberia and Malawi.” National Bureau of Economic Research Working Paper 27932. https://doi.org/10.3386/w27932.