Across the country, there are big variations in the share of employees who report normally working from home, reflecting regional differences in the kinds of jobs that are available. This has profound implications for local economies, including retail, transport and housing.
Prior to Covid-19, fewer than 6% of employees in the UK reported that they normally work from home in their main job. This figure increased to over half the workforce during the pandemic, and it remained at nearly a quarter of all workers in the first half of 2022 (Office for National Statistics, ONS, Labour Force Survey, LFS).
In reality, this may be an underestimate because it does not reflect the large increase – since 2019 – in workers who occasionally work from home. Further, the proportion of workers who are doing their jobs from home is not evenly distributed between sectors or regions of the UK.
Will the transition to home working be permanent?
The pandemic has led to important and seemingly permanent structural changes in how we work. So far, the expected reversal of the massive Covid-19-induced increase in remote working is only partial, and few expect a full return to pre-pandemic office working in the long run.
The UK is not alone in this shift. In the United States, for example, roughly half of paid work hours were done from home between April and December 2020, compared with 5% before the pandemic (Becker Friedman Institute, 2021).
Employers in the United States plan for workers to supply 20.5% of full working days from home after the pandemic ends. Most workers welcome the option to work remotely one or more days per week, according to the Survey of Working Arrangements and Attitudes (SWAA; see Barrero et al, 2021).
There is now substantial evidence to suggest that the increase in remote working – and the corresponding decline in working at the office – will be a permanent shift (De Fraja et al, 2022a; Bloom et al, 2022).
In the UK, recent estimates based on employers’ plans indicate that remote working will be 20 percentage points higher in the foreseeable future than it was in 2019 (De Fraja et al, 2022b). This is roughly in line with the change shown in data from the UK quarterly LFS, and similar to figures for the United States.
A key determinant of whether or not an individual works remotely is the nature of their job (Dingel and Neiman, 2020a; Adams-Prassl, 2020; Dingel and Neiman, 2020b). Job characteristics are also important in determining which occupations will see the largest permanent increases in remote working (De Fraja et al, 2022b).
The biggest growth in remote working tends to be in occupations that already had relatively high rates before the pandemic (see Figure 1). This is important for the regional implications of working remotely in the UK. Specifically, differences in occupational composition across regions lead to variation in the permanent change in working from home.
In 2019, Greater London had 6.7% of workers in health and caring occupations, but 20% in corporate management and business or media, compared with 12.5% and 8.9% respectively in Tees Valley (ONS, 2019).
Because health and care workers have much lower rates of remote working than business and management workers, these labour force composition differences may drive persistent disparities in remote working. This has potentially profound implications for local economies.
Figure 1: Remote working before and after the pandemic by occupation
Source: De Fraja et al, 2022a
How and why does remote working differ across the UK?
Data from the quarterly LFS show substantial regional variation in the proportion of employees who reported normally working from home in the first half of 2022 (see Figure 2).
In London and the South East, around 30-34% reported normally working from home. Comparably, in the North East and Yorkshire and the Humber, this was below 20%. Notice that in 2019, we did not see the same pattern: remote working rates in London were comparable to those in the North of England.
Figure 2: Normally working from home, 2019 compared with 2022
Source: Quarterly LFS, January-June
Note: Values indicate the percentage of respondents who report ‘normally working from home in their main job’.
This regional variation is also reflected in employers’ plans, as reported in the SWAA. For example, in the South West, employers expect a reduction of up to 44% in the amount of work performed at the office, compared with 36% in the West Midlands (see Figure 3).
Figure 3: Expected reduction in work done on business premises by region
Note: The sample includes workers who are able to work from home (as revealed by having done so at some point during the pandemic). From the January 2021 to June 2022 SWAA waves, we obtain employer plans for remote working post-Covid-19. We estimate that the reduction in person days on business premises as the average percentage of full working days from home planned is minus 4.7% (the pre-Covid-19 working from home average based on Understanding Society survey estimates).
Urban areas in England and Wales are expected to see the largest changes in remote working from pre-pandemic levels. Rural areas are likely to see much more modest changes (see Figure 4).
Figure 4: The employer-planned change in remote working for England and Wales
Source: De Fraja et al, 2022a
Note: Values indicate the percentage of work that will be done from home, according to responses to the SWAA question ‘After Covid-19, in 2022 and later, how often is your employer planning for you to work full days at home?’
What are the implications of remote working for regional economies and policy?
The shift to remote working will benefit millions of workers in England and Wales. The average UK employee views working from home two to three days a week as equivalent to a 6% pay raise (Mizen et al, 2021). But the aggregate shift to remote working will affect the economy in several other ways.
Locally consumed services
The change means that many workers are spending more time in the residential neighbourhoods in which they live, and less time in the city centres where they work (De Fraja et al, 2021).
This shift in the geography of where work is done will have consequences for locally consumed services, such as cafes, hairdressers and retail shops. It is estimated that working from home will reallocate £3 billion in retail and hospitality spending from city centres to residential neighbourhoods in England and Wales (De Fraja et al, 2022a). Local authorities in central London will be hit hardest by this change.
Other work finds that, on average, a neighbourhood that experiences a 20% decrease in commuter traffic, due to remote working, will experience a 7% reduction in spending on locally consumed services (De Fraja et al, 2022b). This is in line with recent reports of restaurant closures in city centres and business districts.
There are also reasons to think that the outlets that are forced to close in city centres will not be replaced by corresponding new outlets in the areas where people live and work remotely. These areas, where demand for locally consumed services has increased, are beset by constraints on supply affecting their profitability (De Fraja et al, 2021).
Some constraints could be removed by legislation and active policy, including altering transport routes and easing planning restrictions. Others, such as the population density of the area and the availability of suitable accommodation, cannot be relaxed in the medium term.
Commuting and public transport
Having reduced commuting substantially during the pandemic, working adults in the UK now say that they would prefer to commute just two to three days a week. Further, only one in seven expect to return to commuting five days a week.
These changes imply a substantial reduction in costs and time spent travelling to work. Workers responding to the SWAA spent, in 2019, on average 29 minutes commuting and spent £5.50 per day on travel and parking costs. Over 60% commuted by private vehicle and 34% by public transport (Mizen et al, 2021).
In March and April 2021, the average commuter was travelling four days a week, compared with five before the pandemic, when 10% were not commuting at all. This figure had hardly changed a year later.
When asked about their preferences, working adults in the UK showed a preference for commuting less than they had before the pandemic, favouring just two to three days a week.
Any such reduction in commuting brings other benefits, in terms of faster and cheaper commutes for everyone, including the workers who are unable to work remotely. Traffic management regulations and, in the longer term, infrastructure planning will need to take these changes into account to maximise the positive effects of these externalities.
There is a growing body of evidence about the consequences of increased remote working for housing markets. By reducing the need for daily commuting, a rise in remote working weakens the link between where workers live and where they work.
Empirical evidence supports this idea, finding that the housing market premium paid for living close to the city centre has declined in many large US cities (Gupta, 2021). This has also been seen in Greater London (Gokan et al, 2022).
These findings are consistent with the predictions of theoretical analysis that working from home may reverse trends towards gentrification observed in the last two decades (Althoff et al, 2022; Brueckner et al, 2021; Delventhal et al, 2021).
The consequences for housing markets within the UK could also reach well beyond London. Needing to commute two days per week rather than five makes moving to larger and more affordable housing in smaller cities, villages and rural locations more of an option. Evidence of greater housing price increases in recent years for less productive local authorities in England and Wales is consistent with this behaviour (Gokan et al, 2022).
How might remote working affect the levelling-up agenda?
There is a dearth of research on the effects of these changes and differences between regions on inequalities across the UK, apart from recent use of transactions data (Levelling Up Live 2021). As a result, we can only speculate on the implications for the levelling-up agenda.
Differences in remote working may directly affect regional productivity. Research suggests that adoption of remote working has a small positive effect on firm performance (Harker Martin and MacDonnell, 2012; Bloom et al, 2015). Self-reported evidence suggests a 7% improvement in productivity relative to expectations, but workers claim to be 4% more efficient in the office (SWAA).
These effects are small compared with the differences in productivity between industries. As a result, the largest impact on regional productivity is still likely to be industrial composition by region.
But if firms in relatively productive industries relocate to below-average productivity regions and those workers also work remotely 20% more than five years ago, we might expect productivity to rise in those places and fall elsewhere.
The weakening link between where workers live and work also has the potential to affect the levelling-up agenda by reallocating high-skilled workers from the South of England – where property is relatively expensive – to regions further north – where property is relatively affordable (Brueckner et al, 2021; Gokan et al, 2022).
This reallocation will lead to high-productivity workers continuing to work in high-productivity parts of the country but living in lower-productivity areas.
What further evidence is needed?
The ONS has published data for regional labour productivity at the ITL1 level (for example, East Midlands, Greater London, and Yorkshire and the Humber) in 2020 (ONS, 2020). It is beginning to make regional capital data available at the same disaggregated level and intends to make multi-factor productivity (MFP) estimates – which account for capital inputs as well as labour inputs – available as part of its 2021-23 plan. When these data are available, the ONS should be able to estimate disaggregated/regional total factor productivity (TFP) – a measure of growth in output in excess of growth in inputs (labour and capital).
The ONS has good quality data on housing at the local authority level, allowing an analysis of the impact of remote work on housing. It also has data on commuting from the census. But evidence on remote working patterns is more limited, except through analysis of job vacancy data, the Opinion and Lifestyle Survey and data from the Annual Population Survey. Importantly, none of this is published at the regional level.
Where can I find out more?
- Working from home survey: The Survey of Working Arrangements and Attitudes collects information on current and future working from home from a representative sample of workers in each country. It is led by Jose Maria Barrero, Nicholas Bloom and Steven Davis for the United States, and by Paul Mizen and Nicholas Bloom for its UK counterpart.
- Homeworking in the UK – regional patterns: Using data from the UK LFS, Owain Nolan, James Probert, Nick Chapman, Chris Hendry and Addie Knight document regional patterns in working from home across the UK from January 2019 to March 2022.
- Remote working and the new geography of local service spending: Gianni De Fraja, Jesse Matheson, Paul Mizen, James Rockey and Shivani Taneja estimate the effect that the increase in remote working will have on spending on locally consumed services across England and Wales. They estimate that a neighbourhood experiencing a 20% decline in commuter traffic is expected to experience a 7% decrease in local service spending.
- How the rise in teleworking will reshape labor markets and cities: Toshitaka Gokan, Sergei Kichko, Jesse Matheson and Jacques-Francois Thisse show that teleworking will lead to a movement of skilled workers from high- to low-productivity cities. They find that empirical evidence from housing prices in England is consistent with this behaviour.
Who are experts on this question?
- Brent Nieman, The University of Chicago Booth School of Business
- Jesse Matheson, University of Sheffield
- Paul Mizen, University of Nottingham
- Nicholas Bloom, Stanford University
- Jonathan Dingel, The University of Chicago Booth School of Business
- Henry Overman, London School of Economics and Political Science
- Gianni De Fraja, University of Nottingham
- James Rockey, University of Birmingham