Self-employment has fallen sharply in the UK during the Covid-19 crisis but the extent differs across regions. In the past, it has often risen in recessions and declined in booms. The limited government support for the self-employed may explain the drop during the pandemic.
The number of self-employed individuals dropped markedly during the pandemic. There were over 680,000 fewer self-employed workers across the UK in the third quarter of 2021 compared with the same period in 2019, prior to the coronavirus outbreak. This represents a reduction in self-employment of 13.9% (see Figure 1).
Figure 1: Regional changes in the numbers of self-employed and employees between 2019 Q3 and 2021 Q3, for workers 16 years and older
Source: Office for National Statistics (ONS), quarterly Labour Force Surveys October-December 2019 and 2021
Note: Excludes full-time students
In contrast, the number of employees increased by 1.2% over the same period (see Figure 1). Hence, over the two years, there was a drop in the UK self-employment rate of two percentage points (from 15.5% to 13.5%). This is an abrupt reversal of the previous long-term upward trend in self-employment in the UK, which had peaked at five million prior to the pandemic.
While all parts of the UK experienced a drop in self-employment between the third quarters of 2019 and 2021, there are notable regional variations in the extent of the reduction. This ranged from -2.4% in Wales to -30% in Northern Ireland (see Figure 2). This is striking because there was no significant regional variation in the programmes of support put in place for the self-employed during the Covid-19 lockdowns.
Figure 2: Changes in regional self-employment rates between 2019 Q3 and 2021 Q3, for workers aged 16 years and older
Source: ONS quarterly Labour Force Surveys October-December 2019 and 2021
Note: Excludes full-time students
What might explain these regional differences?
Pre-existing growth in self-employment
Regional variations and growth patterns of self-employment have been relatively persistent over time in the UK (Fotopoulos and Storey, 2017).
This means that broader regional differences remained relatively stable with the highest self-employment rates being persistent in the South of England and the lowest in the North of England, Scotland and the North West (Parker, 2004).
Between the global financial crisis of 2007-09 and the pandemic, regional self-employment rates were generally rising (see Figure 3). Northern Ireland was an exception, where the self-employment rate recovered slowly after the global financial crisis.
Figure 3: Annual changes in regional self-employment rates prior to Covid-19 (2009=100)
Source: Annual Population Surveys from Nomis, 2009 to 2019
In Northern Ireland, self-employment rates have been higher than elsewhere over time, reflecting the size of the agricultural sector in the past. The change may reflect economic restructuring following the 1998 Good Friday Agreement.
London had the highest regional self-employment rate in 2019 at 17.8% of employed workers, but growth there peaked in 2017. By contrast, the North East experienced the fastest growth in self-employment prior to the pandemic, although this was more a delayed self-employment growth (11.9% of employed workers were self-employed in 2019) than a general change in the regional trend of self-employment.
Reasons for the strong continuous increase in self-employment in the UK prior to the pandemic have been much debated (Henley, 2017). For some, it indicates improved opportunities for micro-business start-up activity.
Others have voiced concerns about the rise of the gig economy as a driver of more precarious self-employment that is both necessity-driven and disguised dependent employment (Reuschke and Zhang, 2022; Taylor et al, 2017).
On this basis, a first explanation for the regional variations could be that regions with the greatest increase in self-employment (possibly driven by low-paid gig economy work) prior to the Covid-19 outbreak, may have also experienced the greatest losses in self-employment since then.
Contributing factors may have been that more lower paid employee jobs became available in regional economies as lockdowns ended, particularly in delivery and hospitality (exacerbated by Brexit). Those without a tax record of self-employment in 2018/19 were also not eligible for the Self-Employment Income Support Scheme (SEISS) during the pandemic, which may have driven people back into employment.
But at a regional level, it does not appear that past growth in self-employment explains the decline seen during the pandemic. Higher growth in self-employment between 2010 and 2019 is not associated with a greater reduction in the regional self-employment rate during the pandemic (see Figure 4). If Northern Ireland is excluded, the self-employment rate change during the pandemic is unaffected by pre-Covid-19 growth in self-employment.
Figure 4: Regional changes in self-employment between 2019 Q3 and 2021 Q3 and growth in self-employment prior to the pandemic, 2010 to 2019
Source: ONS quarterly Labour Force Surveys October-December 2019 and 2021, Annual Population Surveys 2010 and 2019
Industries within regions
The industries located in a particular region may have also influenced changes in self-employment. Customer-facing personal services and arts and entertainment were most affected by lockdowns, with more than 80% of businesses having to shut down at the end of March 2020 (ONS).
The experience of these businesses was also more sensitive to the length of the pandemic due to their greater financial vulnerability – for example, expenses and available cash (Bartik et al, 2020).
A second explanation may therefore be that the self-employed in some regions were more exposed to the pandemic than others as a result of the industry structure there. But evidence indicates that, overall, this had only a small impact on the regional variation in self-employment.
This can be shown by measures of regional industrial concentration or specialisation using the number of self-employed workers in 2019 and location quotients that measure the share of an industry in a region to the share of that industry in the UK as a whole. A value greater than one means that there is a over-representation of an industry in a region compared with the UK average. Similarly, a value under one implies under-representation (see Table 1).
Table 1: Regional industry specialisation of self-employment, location quotients, 2019
|Yorkshire and Humberside||1.39||1.17||0.92||1.14||0.65||0.93||1.01||1.00|
|East of England||0.68||0.91||0.99||0.96||0.96||1.16||1.09||0.81|
Source: ONS Annual Population Survey 2019
Note: Standard Industry Classification (SIC); A-Agriculture, BE-Mining, energy, water supply, manufacturing, F-construction, GHI-Wholesale, retail & repair of motor vehicles, transport & storage, accommodation & food services; J-Information & communication; KLMN-Financial & insurance activities, real estate activities, professional scientific & technical activities, administrative & support services; OPQ-public administration & defence, social security, education, human health & social work activities; RSTU – Other services
Pre-existing industrial patterns suggest that there were counterbalancing forces in place that resulted in overall loss in self-employment across all regions. The specialisation scores show that in 2019, self-employment in personal services including arts and entertainment was relatively equally distributed, except from a lower concentration in the East of England and a higher concentration in Scotland.
London’s self-employment was highly specialised in advanced business services – which includes finance and insurance, real estate, professional scientific and technical activities, and administrative and support services, This is a sector that was less affected by measures to control Covid-19 at least initially, due to the high proportion of jobs that could be done from home (Felstead and Reuschke, 2021).
Both London and the South East had high concentrations of self-employment in information and communication. This sector was less vulnerable to lockdowns (ONS). Nevertheless, freelancers may have been at risk of losing contracts even in more crisis-resistant sectors as organisations tried to adapt quickly to the pandemic.
Scotland and Northern Ireland had high specialisation scores in personal services and an under-representation of advanced business services. But they also had a high concentration of self-employment in agriculture, a sector that was less affected by the pandemic.
Wales also had a lower level of specialisation in the advanced business services sectors, and therefore a larger impact on self-employment might have been anticipated. Self-employment is also higher in sectors less affected by lockdown, for example, agriculture and construction. But Wales experienced little growth in paid employment, offering less opportunity for the self-employed to switch.
Employment growth during the pandemic and in the past
Stronger regional economies are associated with declining self-employment rates. Overall, self-employment rates decreased faster in regions where the number of employees grew more between the third quarters of 2019 and 2021 (see Figure 5). This suggests that regions that offered the self-employed opportunities in paid employment have seen a greater decline in self-employment.
Figures 5: Regional changes in the self-employment rate and employee growth between 2019 Q3 and 2021 Q3
Source: ONS, quarterly Labour Force Surveys October-December 2019 and 2021
Stronger regional employee growth during the pandemic is associated with greater regional employment growth pre-Covid-19 (see Figure 6). This shows that regional economies that were stronger before the arrival of Covid-19 also did better during the pandemic.
Figure 6: Regional changes in the numbers of employees between 2019 Q3 and 2021 Q3 and regional employment growth 2016 to 2019
Source: ONS quarterly Labour Force Surveys October-December 2019 and 2021, Annual Population Survey 2016 and 2019
The self-employed were substantially less protected than employees who could access the job furlough scheme. As a result of the pandemic and particularly lockdowns – alongside the consequences of Brexit and Covid-19 infections, which created labour shortages in some sectors – the self-employed might have been ‘pushed’ into paid employment.
Growth in paid employment both during and prior to the pandemic is a significant factor in explaining how self-employment was affected by Covid-19 in UK regions.
The regional concentration of sectors, including of advanced business services in London and the South East, manufacturing in the North and construction in the Midlands, can only marginally explain the regional variation in changes in self-employment.
We do not know if these patterns show a permanent change, or whether exits from self-employment (and entries into paid employment) are temporary, forced by a rather ungenerous support programme and the loss of business opportunities for many sole traders.
Overall, London has experienced a greater increase in jobs during the pandemic and hence provided more opportunities for the self-employed to find alternative employment.
The fall in self-employment may be temporary in the capital and the region may bounce back to previous levels of self-employment when the economy recovers. One study shows that an entrepreneurial path can be restored even after long-term disruptions (Fritsch and Wyrwich, 2014).
In the East and West Midlands, self-employment rates have declined fast. These are regions with lower self-employment rates pre-Covid-19. Self-employment in these regions might recover from the pandemic at a slower rate.
In the North East and in Wales, self-employment has declined less than in other regions. But this is against the backdrop of little growth or even a substantial decline in paid employment and hence opportunities for employment.
This raises the risk of increasing geographical economic inequalities and divergence of regions in the UK that have developed as a consequence of deindustrialisation in previous decades (Agrawal and Phillips, 2020; Martin et al, 2016).
To monitor current regional divergence, both self-employment and paid employment need to be considered – as opposed to overall employment. Including self-employment in the government’s ‘levelling up’ agenda will be important to address uneven economic development and futures across regions (Tomaney and Pike, 2020). Levelling up needs to move beyond dominant beliefs in industrial specialisation.
Where can I find out more?
- Spatial and social impacts of Covid-19 on the self-employed in the UK: Findings from the ESRC project ‘Addressing inclusivity in the spatial and social impacts of COVID-19 on the self-employed in the UK’ (ES/V008781/1) funded by the UK Research and Innovation’s rapid response to Covid-19.
- Time series data on self-employment in the UK: Data series from the Office for National Statistics.
- Regional data on self-employment in the UK: Data from the Annual Population Survey from Nomis.
- Self-employment Income Support Scheme statistics by geography: Data from HM Revenue and Customs.