Did more restrictive mobility policies during the COVID-19 pandemic have lasting impacts on work from home? I argue that changes experienced during the pandemic regarding several work from home arrangements (for example, from school closures and social distancing policies) in states with more strict mobility policies may have become more entrenched for those workers compared to individuals living in states that had more lax restrictions. I examine work from home percentages reported in the annual US Census American Community Survey (ACS) pre and post COVID-19 pandemic to explore whether there were lasting impacts on work from home in states that had more stringent mobility restriction policies versus states with more lax mobility policies during the pandemic. The ACS is the largest US Household survey in the US, with an annual sample size of about 3.5 million addresses. Results from a difference-in-difference estimation show a significant increase in the percentage of workers working from home post COVID-19 in states that had more stringent mobility restrictions than states that had more lax mobility restrictions. The results also show a significant increase in the percentage of workers working from home in states that had longer duration COVID-19 mobility restriction policies compared to other states that had more lax mobility restrictions. These results suggest that workers in more strict mobility policy states became more entrenched with their work from arrangements and that COVID-19 public health policies had lasting effects on remote work trends in the US. Overall, this paper explores how public health policies during the COVID-19 pandemic influenced remote work trends in the US.

Are stay-at-home orders effective during a global pandemic? Although stay-at-home orders should help to slow the spread of contagious diseases (like COVID-19) by reducing person-to-person contact outside a household, these orders are only effective if people actually stay at home. This study uses data on the mobility of (anonymized) smartphones within states before and after the enactment of stay-at-home orders to understand the effects of stay-at-home orders on mobility. The dataset contains over ten million observations on the movements of smartphones across all states within the US, which allows for an analysis of compliance with stay-at-home orders without worrying about the potential for self-reporting biases. By using a quasi-experimental method, this study overcomes biases that can come from comparing pre and post policy trends (due to unmeasured differences across states). The results from a difference-in-difference analysis suggest that stay-at-home orders are associated with a 4.6% increase in the percent of smartphones that remained at home during the late spring of 2020 across the US. Although there is a statistically significant difference across states with and without stay-at-home orders, it is important to note that the average percent of smartphones that remained at home was 40% in stay-at-home states compared to 36% in states without stay-at-home orders. The results also show that penalties (jail time and fines) had no significant effect on compliance with stay-at-home orders (considering all states with such orders), while compliance with stay at home orders in Republican controlled states was 5.3% lower than in Democratic controlled states. Further, states with higher population densities had the highest percent of smartphones that remained at home after stay-at-home orders went into effect (at 45%). Overall, the results in this paper suggest that stay-at-home orders had a small but significant impact on mobility, while also suggesting that studies of individual behaviors and choices will be necessary to understand when and why people may be more or less willing to shelter at home during a global pandemic.

Although several studies have examined the spread of information based on the positive or negative content of information, there are many human emotions that are difficult to categorize into positive or negative categories. This study examines the diffusion of information considering six sentiment categories (including happiness, sadness, anger, fear, surprise and disgust). It is argued that there will be differences in the diffusion rates of tweets considering six different sentiments and four different topics (including climate change, 2020 elections, gun control and crowdfunding). The results by topic reveal the importance of examining differences across topics. For both the 2020 election and gun control topics, angry tweets are significantly more likely to be retweeted than all other sentiments. However, for both the crowdfunding and climate change topics, both happy and angry tweets are equally likely to be retweeted. Further, happy tweets are significantly more likely to be retweeted than tweets with sad, fear or disgust sentiments for both the crowdfunding and climate change topics, while there were no significant differences in the diffusion of these sentiments in the 2020 election and gun control topics. All of the results in this paper are robust to dropping outliers and to dropping retweets by bot accounts. Overall, the results suggest that the broad categories of positive and negative sentiment are too aggregate to really explain when information is likely to be diffused through retweets.

Several government agencies (including the CDC and other federal and state level agencies) have been collecting and creating big datasets around the COVID-19 global pandemic. These data provide important information that can be used to inform our understanding of the effectiveness of different policies and measures in reducing the spread of the virus. In this paper, we explore the relationship between mask mandates and COVID-19 infection rates to try to understand the role that mask mandates play in infection growth rates. Using a difference-in-difference estimation and data from the CDC and other state agencies, the results suggest that mask mandates are associated with a 15% decrease in COVID-19 infection growth rates during the late spring/early summer of 2020. These results are not impacted by differences in population density. Considering Republican controlled states only, masks were associated with a 9% decrease in COVID-19 infection growth rates. Considering states that had stay-at-home orders only, mask mandate were associated with an 18% decrease in COVID-19 infection growth rates. Overall, the results in this paper suggest that mask mandates had a significant impact on reducing the spread of COVID-19 during the late spring and early summer of 2020 in the US.