Harvard Talks: Work From Home During Pandemic

Work From Home

To read the full report of Prof. Prithwiraj Choudhury, Harvard Business School, click here

Prior research has documented that during mortality-related crises workers face psychic costs and are motivated to make social contributions. In addition, management practices that encourage workers to make social contributions during a crisis create value for firms. However, the coronavirus crisis of 2020 is unprecedented given conditions of social distancing.

Work from Home vs Cannot Work from Home

It raises the question of whether workers who continued to work (albeit from home) during this crisis were constrained in their ability to make social contributions and exhibited disproportionately greater psychic costs compared to workers who could not work from home. We exploit this shock to estimate differences in content contributions to an online community by workers who work from home (WFH) relative to workers who cannot work from home (CWFH).

Online content contributions are especially pertinent in our context because social distancing constrained traditional forms of social contributions such as physical volunteering. Using data from a popular question-and-answer platform, we estimate a difference-in-differences specification and report nuanced results: while Work From Home workers made 19% fewer online contributions on average and contribute less to topics such as ‘family’, they make 148% more contributions on topics related to ‘Work From Home best practices.’

Using natural language processing tools, we also find that Work From Home workers exhibited greater psychic costs than CWFH workers. We provide evidence for a plausible mechanism, i.e. time allocation, and show that Work From Home workers attempted to catch up on social contributions at the end of their workday, suggesting time constraints. Our research contributes to literatures on managing workers during a crisis, Work From Home and online communities, and have several immediate implications for managing Work From Home and CWFH workers during the coronavirus crisis.

Social Distancing Shock

The coronavirus shock of 2020 is one of the most acute crises affecting companies and workers in recent times. In prior literature, Carnahan et al. (2017) highlight that a mortality related crisis (in their paper, the 9/11 shock) leads to psychic costs for workers and motivates them to make pro-social contributions. In addition, management practices that encourage workers to make social contributions during a crisis create value for firms.

In fact, prior literature in strategic human capital has stressed that firms should motivate workers by offering them opportunities that increase worker satisfaction (Gambardella et al. 2015) and this insight is deeply relevant during a major crisis. However, the coronavirus crisis is unprecedented in that conditions of social distancing forced companies around the globe to ask millions of workers to work from home (Bodewits 2020). This created variation in work arrangements across industries as workers in some industries cannot work from home and raises the question of whether workers who continued to work (albeit from home) during this crisis were constrained in their ability to make social contributions.

We also ask whether workers who made the drastic switch to working from home exhibited greater psychic costs and were constrained for time in the immediate aftermath of the crisis, compared to workers who could not work from home.

Social Contributions Affected

Mandatory social distancing not only created variation in work arrangements for workers, it also constrained traditional forms of social contributions such as physical volunteering. However, workers could still make social contributions, albeit online. In today’s digital economy, an important form of social contribution by workers relates to the unpaid content they contribute to online communities (Boudreau and Lakhani 2013; Zhang and Zhu 2006, 2011; Ghose and Han 2011; Luca 2015). Such content represents an example of a “positive spillover” from the internet to society as discussed by Chan et al. (2013). To quote Zhang and Zhu (2011), “Many public goods on the Internet today rely entirely on free user contributions. Popular examples include open source software development communities (e.g., Linux, Apache), open content production (e.g., Wikipedia, Open Courseware), and content sharing networks (e.g., Flickr, YouTube).” (ibid, page 1601). Thus, we adopt a broad definition of social contribution to mean any free contribution to an online public good.

We exploit the coronavirus shock and provide causal evidence of how workers who work from home (WFH) differ from workers who cannot work from home (CWFH) with respect to content contributed to online communities. The online community of interest is one of China’s largest online question and answer communities. In prior literature, Wang et al. (2013) study online question and answer communities such as Quora and Stack Overflow and conclude that “community question and answer sites provide a unique and invaluable service to its users” (ibid, 1350). Thus, our measure of individual worker output (i.e., the number of answers contributed to the online community) meets the characterization of social contributions (i.e. “helping others”) articulated by Carnahan et al. (2017).

Our pre-period comprises the Chinese New Year holidays when most workers are on vacation in their hometowns. In the post-period, all workers were under mandatory lockdown facing social distancing. However, while individuals working in certain industries (such as the internet industry and the software industry) were asked to WFH, workers in industries such as airlines, computer hardware, and manufacturing could not work from home under lockdown.

We estimate a difference-in-differences specification comparing content contributions of individuals working from home in the post-period to contributions made by individuals who cannot work from home in the post-period.

WFH Report

We report several results.

On average, WFH workers posted 19% fewer answers than CWFH workers. However, WFH workers provided more answers in certain socially helpful categories such as advice on how to work from home.

Using natural language processing tools, we also document evidence of greater psychic costs exhibited by WFH workers compared to CWFH workers. Additionally, we explore a key mechanism (i.e. time allocation) and analyze hourly contributions by WFH and CWFH workers.

This reveals that while during work hours WFH workers contributed less than CWFH workers, they contributed equally between 8pm and midnight.

In other words, WFH workers tried to “catch-up” on the social contributions at the end of the workday, suggesting disproportionate constraints on their time. Our findings contribute to several streams of the strategic human capital literature and have implications for how managers and firms should manage WFH and CWFH workers during a major crisis.

To access the full report of Prof. Prithwiraj Choudhury, Harvard Business School, click here