In the wake of COVID-19, we saw and continue to see global disruptions; disruptions such that livelihoods and lives were being uprooted, production processes and supply chains were at a halt, and populations at the margin kept bearing the brunt of the crisis. A significant chunk of us, privileged members of society, began to witness the stark differences between our personal experiences and how a majority of the population, especially in low and middle income countries has to suffer. I, for one, began to understand the specific nature of social disparities and the urgency with which they needed to be tended to. Privilege, I realized, was a relative term – people from different walks of life were assigning different utilities to roughly similar experiences. It was in the midst of a pandemic that we saw many people pay attention to these problems which were in fact, always right in front of us.
To paint a more targeted picture from the context of my country, almost 25% of Pakistanis live below the national poverty line, with an additional two-fifths identifying as multidimensionally poor. An estimated 20-30% of the population, experiences food insecurity in one form or the other. Alarmingly, FAO’s estimates from 2020 presented a picture of ~36 million people being highly and persistently exposed to natural disruptions and being chronically vulnerable to food insecurity. With such crucial estimates, targeted policymaking becomes a need of the hour. As a starting point, Pakistan as an agrarian economy experienced uncertainties in food production, supply and demand of essential food items due to a lack of affordability, soaring prices, etc during initial lockdowns. Income and employment losses not only served as losses of livelihoods, but for people at the peripheries, they served as losses of lives.
Getting NGOs and interested public-private stakeholders on board to allocate resources towards identifying purposive samples of at-risk families would suffice as a decent starting point. Real-time data on impacts and vulnerable populations is needed immediately, not only in Pakistan but in other developing and low-income countries. Without availability of timely data, emergency interventions are likely to lack proper design and/or targeting. Hence, collecting data on food-insecure households, especially in rural areas where most of the population suffered the adverse effects of soaring food prices could make it possible to uncover information that can help us direct efforts to help them. Among the most vulnerable groups are women and children, who are likely to have suffered because of higher food prices. If nutritious food items are expensive because of disrupted supply chains and production processes, maintaining dietary diversity among the poor and most vulnerable becomes very important, especially for children under-five. Collecting regular and timely data could help put social safety nets in place for future emergencies, and simultaneously identify and develop behavioral nudges that have spillover effects on direct/indirect influencers for diverse population groups.
As a young economist keen on integrating behavioral policymaking, I can’t help but go back to an experimental evaluation I was a part of. In an attempt to investigate the decision-making abilities of a sample of randomly chosen students, I used maximum likelihood estimation, a statistical technique that helped obtain realistic estimates for human behavior. The evaluation framework provided a mechanism for machine learning, framing each student’s decision in the face of uncertainty as an optimization problem. Under set beliefs, it showed that rational students behave in ways that depict their true preferences. These findings were critical – while we expected students with higher scores to exercise caution in high-stake situations, their self-perception suggested otherwise; students scoring high in our test of ability were willing to indulge in risky behaviors, despite being at odds with societal norms. Given our findings, I wondered how similar frameworks could extend on to the potential collateral effects of the ongoing health crisis. With models estimating increases in wasting (low weight-for-height) rates to be somewhere between 10 and 50%, were parents aware of the irreversible damage childhood stunting causes? Was compliance to social norms more valuable than their child’s health? Could innovative, human-centered emergency preparedness systems help protect populations at risk of malnutrition from also suffering from food-insecurity? These questions sparked my passion for working with behavior-led frameworks instead of traditional approaches to intervention design. Since research in Pakistan has often overlooked psycho-social determinants of dietary choices, my solution takes a more data-oriented approach to explore the underlying phenomena that shape attitudes, perceptions, and feeding practices of mothers and influential household members in underserved areas. An approach that may have success is mobilizing field teams and community health workers to collect data and also educate mothers, pregnant and lactating women, etc. on community-specific, culturally appropriate feeding practices. This will not only lead to improvements in the overall wellbeing of the family but also increase buy-in from influential family members’ who may otherwise be opposed to a healthy practice but are willing to listen to a credible community resource.
As an agrarian economy, such data collection can also help by supporting farmers in low-income areas to engage in production of palatable, culturally appropriate therapeutic food products that have been critiqued for a lack of uptake, and subsidize production of fortified food items for complementary/supplementary feeding. At the same time, if planned and designed properly, we could see a ripple effect by way of creating employment opportunities in the farming/agricultural sector, preventing future productivity losses for children, and a reduced burden of the eventual collateral impact on food production and access, provision of health services, and improve effective practices and behavior in lieu of natural emergencies and hazards. Ultimately, it becomes important to direct resources towards regular and timely data collection, build comprehensive data reserves that help direct efforts where they are needed and use this opportunity to build back better in a way that puts communities and people at the core. The collateral impact of the pandemic can only be reduced by efforts being redirected towards innovative emergency warning systems that help reimagine the landscape of low-income economies.
1411-11