Covid-19 briefing: working from home and worker productivity

John Lewis, Andrea Šiško and Misa Tanaka

The Covid pandemic has led to a large enforced shift towards working from home (WFH) as a result of ‘stay-at-home’ policies in many countries. This led to a resurgence in interest in, and new reignited discussion about, the consequences of greater WFH. In this briefing we review the literature on the impact of WFH on productivity. Across a very diverse literature the key lessons are: impacts depend on the nature of tasks, the share of WFH matters, and there is big difference between enforced versus voluntary WFH. And the caveats are important too: cost savings at the firm level don’t automatically translate into economy-wide productivity gains and evidence on long-run effects remains very scarce.

The literature on the topic is large and varied. In this post our focus is on the effect on productivity. We therefore abstract from broader issues such as wellbeing, management style, labour markets, spatial implications, sectoral shifts and other potential impacts.

Effects of WFH on productivity differ significantly across tasks and roles

Perhaps the most well known paper in the literature is Bloom et al (2015). This conducts an experiment to study the impact of WFH on the performance of Chinese call centre workers responsible for airfare and hotel bookings. Workers who volunteered to WFH were randomly assigned to WFH or work in office (WIO), to safeguard against sample selection bias effects. Those assigned to WFH had a 13% performance increase relative to those who were assigned to WIO. Some of this increase is attributed to taking fewer breaks and sick days, and some to quieter working environment which enabled workers to take more calls per minute.

But other studies find that physical separation of workers can reduce productivity for other types tasks, eg when teams need to work together to resolve urgent and complex tasks. Battiston et al (2017) exploit a natural experiment involving 999 call handlers and radio operators in the United Kingdom. They find that performance – measured by the time taken between the creation of the incident by the call handler and the allocation of police officers by the radio operator – is 2% better when teammates are in the same room, and attribute this gain to the benefits of face-to-face communication.

Golden and Gajendran (2019) also find evidence that the impact of WFH on productivity depends on the role. They use matched survey data of employees who WFH voluntarily and their supervisors in an organisation. Overall, the authors find a positive relationship between WFH and job performance. But there was a stronger positive association between performance and the extent of WFH in roles with greater job complexity and less interdependence.

The relationship between WFH and workplace productivity may be non-linear

Using five case studies in two large telecommunication firms, Coenen and Kok (2014) find that WFH improves the speed and quality of new product development, provided that face-to-face contact, which is more likely to lead to trust building and high-quality knowledge sharing, is not completely replaced by virtual contact. Kazekami (2020) studies the influence of WFH on worker productivity in Japan using survey data in 2017 and 2018 and finds that the relationship is hump-shaped: some degree of WFH boosts productivity, but when WFH hours are too long, productivity falls. A meta-analysis of 46 psychological studies by Gajendran and Harrison (2007) also suggests that more than 2.5 days/week of WFH could harm relationships with co-workers. To assess the long-run impact of WFH on GDP, Behrens et al (2021) build a general equilibrium model. They find that 1–2 days (20%–40%) of WFH would maximise the long-run GDP given the current level of ICT development, as higher levels of WFH would reduce productivity.

Home environment relative to office environment is a key determinant of the productivity effects of WFH. Homes are not primarily designed for work and their suitability differs across workers depending on their financial means and living situation. ILO (2020) notes challenges both in setting up an appropriate home-working conditions and in monitoring and controlling compliance with occupational health norms. PWC (2020) points to lack of physical space, privacy and inadequate technology as potential challenges, which may be greater for younger workers (who are more likely to live in smaller dwellings and in households with more working adults) and those who have children at home. Evidence during Covid lockdown also suggests that working while having children at home is productivity reducing, particularly for mothers. Andrew et al (2020) conduct a survey of UK two-parent households with children aged 4–15 during the school closure period of April–May 2020, and find that mothers were doing only a third of uninterrupted paid-work hours of fathers on average. 

But some studies have also shown that certain types of office designs can reduce productivity and collaboration. Smith-Jackson and Klein (2009) study the effect of background noise on task completion rates and find negative consequences of noisier open spaces on performance. Bernstein and Turban (2018) conduct a study of bilateral interactions measured by sociometric devices and find that more openness in offices lowers face-to-face interactions by 70%, as employees sought more privacy. Interestingly, Coenen and Kok (2014), discussed earlier, find that the incidence of WFH increased after hot desking was introduced. A survey-based study by Oxford Economics (2016) points to ability to focus without interruptions, device connectivity, having collaborative areas, and giving employees their own space as the most important features for office design to enhance productivity.

Choice matters

The Bloom et al (2013) study discussed earlier found that about 50% of the initial volunteers changed their minds and decided to work in the office after the end of the experiment, citing social reasons and that it was troublesome for other household members. Around 10% of the initial non-volunteered group opted to work remotely because they saw the success of their peers who worked from home.

Palumbo (2020) and Oakman et al (2020) tended to find that freedom to choose where to work and the perception of autonomy are important factors in explaining the positive outcomes from remote working. But enforced WFH, driven by the desire to cut office space, could lead to a different outcome. For example, Bloom (2020) notes that his oft-cited study drew on a case where employees volunteered to WFH and were only allowed to if they had an office at home where no one else was permitted to enter. He describes enforced WFH in unsuitable spaces as a ‘productivity disaster for firms’.

Broader considerations

Studies on WFH tend to focus on narrower, short-term outcomes rather than longer-term effects. Existing empirical and experimental studies typically look at a narrow, well-defined set of tasks and tightly specified productivity metrics. That literature is relatively silent on longer-term effects and effects that are difficult to measure, such as innovation, employee retention, integration of new colleagues, and team cohesion. 

In addition, current WFH experiences may be affected by the stock of social capital built up from earlier office-working experiences in terms of networks, institutional knowledge and trust building (Haldane (2020)). If shifts to greater WFH affect social capital build-up, the effects may be different to current experiences. Equally, business models and IT may not (yet) have adapted to a world with greater WFH.

It is also important to note that making workers WFH can result in cost-saving by individual firms, but not necessarily increase productivity in the aggregate economy. Measured productivity relates aggregate output to capital and labour inputs. Greater WFH could enable firms to save costs on intermediate inputs and land (office space, heating, lighting etc). But if this simply shifts costs from employers onto workers whilst keeping the underlying ‘production function’ in terms of labour and capital inputs unchanged, it will have no direct effect on productivity. The individual firms’ decision to shift their workers to WFH could result in aggregate productivity gains only if workers can be more productive at home rather than in office, or if companies use WFH to cut office space without damaging their own productivity and the ‘freed-up’ space is then used by others for alternative productive purposes. 

Closing thoughts

A recent survey by Taneja et al (2021) suggests that nearly 80% of UK workers would prefer to work from home at least a few days a week post Covid. It remains to be seen whether the sudden and rapid rise in WFH leads to more permanent changes in working patterns and styles. The diverse literature on WFH has only grown since the pandemic started and debate is likely to persist. The key messages from the work so far suggests heterogeneity in impacts with respect to tasks and workers, the relative quality of home versus office environments matters and that the enforced versus choice dimension is important. But given the rich environment for future studies provided by Covid, and the heightened interest in the topic over the past 12 months, the existing body of literature is likely to be only the start of a much longer and broader-ranging discussion.

John Lewis, Andrea Šiško and Misa Tanaka work in the Bank’s Research Hub.

If you want to get in touch, please email us at or leave a comment below.

Comments will only appear once approved by a moderator, and are only published where a full name is supplied. Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.