Diagnosis of COVID-19 impact on the construction employment: Decline and recovery

Abstract

The Coronavirus Disease 2019 (COVID-19) had an unprecedented impact on U.S. construction employment. The assessment of employment declines and recoveries across various construction sectors, workforce demographics, and geographic regions helps develop inclusive long-term plans to overcome the setbacks of natural hazards, such as the COVID-19 pandemic. Existing methods for quantifying the employment decline from sudden shocks do not consider trends and seasonality of construction employment under normal conditions. Hence, it is not clear whether the employment declines are due to the impact of a disaster or associated with trends and seasonal patterns of employment. The objective of this research is to develop an approach based on time series models, i.e., seasonal autoregressive integrated moving average (SARIMA) models, and cumulative sum (CUSUM) control charts to statistically quantify the impact of the COVID-19 pandemic on construction employment. Seasonal ARIMA models are developed using the pre-pandemic employment data from January 2011 to December 2019, and employment estimates under normal conditions are projected for the period from January 2020 to August 2021. CUSUM control charts are then used to detect employment declines and recovery timeframes for different construction employment. Results show declines in all construction sectors, faster recovery for women and Hispanic workers, and residential building jobs rebounding first. By August 2021, 14 states recovered construction employment, but 36 states lagged. It is anticipated that construction workforce decision-makers can benefit from this study by enhancing their understanding of the employment declines and the recovery status across industry sectors, gender, race, and geographical regions.