Forest-Based Employment in the Southern United States amidst the COVID-19 Pandemic: A Causal Inference Analysis

Abstract

Abstract In response to the COVID-19 pandemic, some governments instituted mandatory stay-at-home policies. As these policies made exceptions for essential industries such as the forest sector, it is not clear a priori whether and how these policies would affect forest-based employment. This study examined the effect of mandatory stay-at-home orders on employment in the forest sector in forest dependent counties in eleven southern states in the United States. We estimated panel event study models in addition to difference-in-difference models to evaluate the policy effects with respect to initial treatment as well as average effects to the treated counties during treatment. We found that employment in wood product manufacturing was lower on average in counties under a lockdown order compared with employment in counties without an active order. We also found that employment in the aggregate forest sector in counties that implemented the policy was significantly lower than employment in counties without the policy up to 9 months following initial enactment compared with the underlying difference between treated and control counties one month prior to treatment. Overall, our findings suggest that lockdown policies had a limited effect on employment in essential industries such as wood-based manufacturing sectors in the southern states. Study Implications: This study uses causal inference techniques to assess the effect of lockdown orders on forest-based employment in 11 southern states. Results suggest effects of lockdown orders were limited to the aggregate forest sector and wood product manufacturing, likely due to the rapid designation of the sector as essential and the reduced production of sawmills. The other industries within the sector were not significantly affected, implying that industries designated as essential may be more resilient to short-term shocks. These findings can be used to inform discussions on potential policy responses to future disasters, including how different interventions interact with one another.