Working from home can save energy and reduce emissions. But how much?

As the Covid-19 crisis spread around the world, large numbers of people started working from home, with immediate and varied impacts on energy use. Oil demand shrank but residential electricity use surged. Companies such as Google and Facebook announced they would allow staff members to work remotely until at least the beginning of next year, while Twitter said its employees could continue working from home indefinitely.

This raises the question of what the implications would be for energy use and greenhouse gas emissions if a significant amount of people continued regularly working from home in the years to come. 

Our analysis shows that for people who commute by car, working from home is likely to reduce their carbon dioxide (CO2) footprint if their journey to work is greater than about 6 kilometres. However, for short car commutes or those done by public transport, working from home could increase CO2 emissions due to extra residential energy consumption.

By analysing commuter trends and labour market data, we found that if everybody able to work from home worldwide were to do so for just one day a week, it would save around 1% of global oil consumption for road passenger transport per year. Taking into account the increase this would bring in energy use by households, the overall impact on global CO2 emissions would be an annual decline of 24 million tonnes (Mt) – equivalent to the bulk of Greater London’s annual CO2 emissions. 

This is a notable decline but small in the context of the emissions reductions that would be necessary to put the world on a path towards meeting key long-term sustainable energy and climate goals. If everyone who can work from home were to do so more frequently than one day a week, the reduction in emissions would most likely be proportionally larger. However, a significant and sustained shift towards working from home could have impacts elsewhere in the energy system, such as those related to preferred modes of transport and demand for office space.

The Covid-19 crisis has had staggering consequences for the transport sector. Government lockdowns triggered a fall of 50% to 75% in road traffic around the world. In April, with around one-third of the global population in complete lockdown, gasoline use dropped by more than 9 million barrels a day – an unprecedented fall – and demand for diesel was down by 6 million barrels a day. With up to 59% of employees in affected countries working from home and some facing redundancy, the effect on rush-hour road traffic was even more striking. Major cities saw a drop in rush-hour congestion of 65% to 95%. There were also widespread declines in air pollution from road traffic. One of the steepest was in New Delhi, where average levels of nitrogen dioxide were around two-thirds lower during lockdown compared with the weeks leading up to it.

Lockdowns have also affected residential demand for energy. Although overall electricity consumption plunged by 20% or more, energy utilities reported increased residential demand as a result of people spending more time at home. Hourly demand patterns on weekdays resembled those of a normal Sunday. In some parts of the United States, average residential electricity use on weekdays was up by 20% to 30%. In the United Kingdom, residential electricity consumption jumped by 15% in the days after the lockdown began. 

There is no guarantee that car use will remain low in the immediate aftermath of lockdowns. Prompted by perceived health risks, the shift away from public transport could continue as demand for mobility returns to normal, leading to a rebound in oil consumption. A survey in China by market research firm Ipsos reported a 57% reduction in the share of journeys made by bus and metro, but a doubling in the share made by private car. Globally, if 10% of bus trips were to be made by car instead, this would add some 700,000 barrels a day to fuel demand for cars – roughly 3% of the total amount of oil used for passenger road transport in 2019.

Average rush-hour traffic congestion in selected cities in 2019 and during lockdowns

Openexpand

However, the lockdown experience could give those who can work from home an increased appetite for it, leading to a lasting reduction in commuter traffic as well as a possible increase in residential energy consumption. 

If the overall level of working from home in an economy remained low, then almost all of the drop in transport energy demand would come from the commuter trips by car or motorcycle that would be avoided, since buses and trains would continue to run – albeit at slightly reduced capacity.

Average change in energy demand and CO2 emissions from one day of home working for a single household with a car commute

Openexpand

The impact of home-working on transport varies widely depending on the region and the time of year. In the United States, the average one-way commute by car is around 18 kilometres, and over three-quarters of car commuters travel alone, according to the US Census Bureau. In Europe, the average one-way car commute is 15 kilometres, and in China it is 8 kilometres, with large variations between urban and rural commutes. Differences in fuel efficiency also matter, as the average car in the United States consumes around 45% more fuel than the average in Europe for a trip of the same length. 

The use of mobile air conditioning in cars also has a material impact on fuel consumption, ranging between 3% of overall annual consumption in colder climates to 20% in hotter ones. Mobile air conditioning can peak at over 40% of fuel consumption in warm climates and congested traffic. We estimate that around 4% of total fuel consumption for commuting by car in the United States, China and Europe in 2019 was for mobile air conditioning.

On the residential side, a day of working from home could increase household energy consumption by between 7% and 23% compared with a day working at the office, depending on regional differences in the average size of homes, heating or cooling needs and the efficiency of appliances. In most parts of the world, the extra demand in winter is larger than in summer, due to space heating, and the energy mix in winter typically shifts more towards fossil fuels. In the US, however, the widespread use of air conditioning results in higher electricity demand in summer than in winter. In China, the prevalence of district heating – which is likely to remain switched on irrespective of whether a household is occupied or empty during the day – reduces the energy impact of working from home in the winter.

For commuting by car, a day working from home would on average reduce energy consumption and CO2 emissions. However, for short commutes by car (less than 6 kilometres in the United States, 3 kilometres in the European Union, and 2 kilometres in China), as well as for those made by public transport, working from home could lead to a small increase in emissions as a result of extra residential energy use. The net effect would depend on the residential fuel mix. For instance in China, the net drop in energy demand translates into only a small decrease in CO2 emissions – especially in winter – because of the relatively high emissions intensity of the power sector as well as the widespread use of coal for space heating and oil for water heating and cooking. 

At the household level, the impact of working from home on energy demand varies widely according to many factors related to the season and the region. But what would be the global effect if working from home became a trend for societies at large? The answer to this question depends in part on how many people can work from home.

Before the Covid-19 pandemic, an estimated 8%1 of the global workforce was working exclusively or mainly from home, with large differences among countries, according to the International Labour Organization. For example, around 5% of workers in China were working from home, compared with 14% in the Netherlands.

Based on an analysis done by Jonathan Dingel and Brent Neiman at the University of Chicago as well as work by the International Labour Organization and others2, we estimate that around 20% of jobs globally could potentially be done from home. This ranges from around 10% in sub-Saharan Africa to more than 45% in the wealthiest European countries. Overall, there is a positive correlation between the potential to work from home and GDP per capita. This reflects the differences in economic and occupational structures of countries, as well as digital readiness (e.g. broadband internet access, computer ownership), and other structural factors (e.g. housing situation, other types of home-based work).

Change in global CO2 emissions and final energy consumption by fuel in the “home-working” scenario

Openexpand

Working from home will normally reduce net energy demand for a household that commutes by car. But for commuters taking public transport, it is likely to increase net energy demand, although regional and seasonal differences are significant. Taking this into account, however, we find that during an average year, the overall energy saved as a result of less commuting is still around four times larger than the increase in residential energy consumption.

On a global level, oil savings3 are around 11.9 million tonnes of oil equivalent (Mtoe) per year – or about 250,000 barrels a day, corresponding to about 1% of road passenger transport consumption. After including the extra residential demand4, overall energy use falls by around 8.5 Mtoe, resulting in a drop of 24 Mt in annual CO2 emissions. 

The scenario described here assumes a relatively modest one day per week of working from home. If this were to become more frequent, the impacts on private transport and residential energy demand would increase proportionally, but it could also have an increasing number of effects elsewhere in the energy system. Given that only one-fifth of workers can do their jobs remotely, even several days of home working per week would probably have only a small direct impact on public transport. However, fewer commuters could lead to less congestion, faster moving traffic and a possible shift away from public transport. This could either reduce or increase oil demand depending on the specific circumstances.   

The longer term impacts on energy and emissions of a trend towards greater working from home are uncertain. Over time, a more significant shift to home working could also result in a reduction in demand for office space and energy for commercial buildings, and therefore a greater overall reduction in energy consumption and CO2 emissions. However, habitual home working could lead to people living farther from their place of work, potentially offsetting the demand reductions in energy for commuting.


Additional contributions to this commentary were provided by Apostolos Petropoulos, George Kamiya, Yannick Monschauer and Kevin Lane.

References
  1. This includes employees who telework, as well as a wide range of occupations including industrial outworkers, artisans, self-employed business owners, and freelancers. Among employees, about 3% were working exclusively or mainly from their home before the Covid-19 pandemic. Source: ILO (2020).

  2. Boeri et al. (2020); Brynjolfsson et al. (2020); Eurostat (2018); Saltiel (2020).

  3. These savings are estimated by combining the country-level potential for home working with employment data and regional characteristics of commuter travel, as well as average commute distances and vehicle efficiencies.

  4. The potential increase in residential energy consumption is based on combining a regional bottom-up analysis of residential demand by end-use and fuel with week-end and weekday electricity load curves, the emissions intensity of electricity and employment data from the International Labour Organization.