COVID-19 Lockdown & Technology Engagement: New Evidence from a Large-Scale m-Health Intervention in India

Abstract

Information provision for social welfare via cheap technological media is now a widely available tool used by policy makers. Often, however, an ample supply of information does not translate into high consumption for information due to various frictions in demand, possibly stemming from pecuniary and non-pecuniary cost of engagement, along with institutional factors. We test this hypothesis in the Indian context using a unique dataset comprising two million call records of enrolled users of NGO ARMMAN that sends timely informational calls to mobile phones of underprivileged pregnant women. The strict lockdown induced by COVID-19 in India was an unexpected shock on the opportunity cost of engagement with m-Health technology, both in terms of reductions of market wages and increased time availability at home. Using a difference-in-differences design on unique calls tracked at user-time level with fine-grained time-stamps on calls, we find that during the lockdown period, the call durations increased by 1.53 percentage points. However, technology engagement behavior exhibited demographic heterogeneity increasing relatively after the lockdown for women who had to borrow the phones vis-à-vis phone-owners, for those enrolled by direct outreach programs vis-à-vis self-registered women, and for those who belonged to the low-income group vis-à-vis high-income group. Our results have policy relevance around demand-side frictions for technology engagement in developing economies and gender implications herein.