Self-Reported Food Insecurity and Objective Poverty Measures in Periods of Subsequent Shocks: What Do Subjective Indicators Capture? The case of Mozambique

Accurate measurement and continuous tracking of household food security are central to the design and evaluation of poverty-focused development interventions. Food security is a core dimension of household poverty and vulnerability. Research on food security and poverty has traditionally relied on objective indicators such as household consumption, calorie availability, and multidimensional poverty indices. However, a growing body of literature highlights the value of subjective and self-reported measures of food insecurity, which may capture dimensions of deprivation and vulnerability not reflected in standard welfare metrics. This study examines self-reported food insecurity indicators in Mozambique and compares them with conventional welfare measures over the period 2014/15–2022, a decade marked by successive and compounding shocks. Using data from the 2014/15, 2019/20, and 2022 household budget surveys, the analysis shows that the determinants of self-reported food insecurity extend beyond household consumption levels, calorie availability, and multidimensional poverty status. Moreover, for 2019/20 and 2022, the surveys include information on both objectively measured and self-reported shocks, as well as access to social assistance programmes. We find that, while objectively measured exposure to shocks is almost never associated with subjective food insecurity, self-reported exposure to shocks is more strongly correlated with these indicators. Also, receipt of social protection benefits neither appears to improve households’ reported food security, nor does it mitigate the perceived impact of shocks. In terms of temporal patterns, trends in self-reported food insecurity closely mirror national consumption-based poverty estimates. Self-reported food security indicators deteriorate sharply between 2014/15 and 2019/20, followed by a moderate improvement between 2019/20 and 2022. This is consistent with the trajectory of consumption poverty over the same period. However, these patterns vary across sub-populations. Overall, while the findings validate the direction of official poverty estimates, they offer a more nuanced understanding of households’ experiences of food insecurity. They underscore the value of integrating subjective food insecurity indicators into poverty-monitoring systems to better capture the challenges faced by vulnerable households in low-income settings.