VoxEU Column COVID-19 Labour Markets

The large and unequal impact of COVID-19 on workers

The spread of COVID-19 has already had a large negative impact on labour supply and earnings of workers in many countries. In this column, the authors leverage newly collected data from the US and the UK to show that these negative consequences are particularly harsh for younger workers, those with unstable employment relationships and lower labour income. The evidence calls for a quick response from governments in the form of stimulus and labour income replacement packages, and a robust plan to ensure that the younger generation are not permanently disadvantaged.

The world economy is enduring a staggering downturn amidst the spread of COVID-19. Individuals are already suffering immediate losses in terms of income and employment. Economists have been quick to call for governments to take strong coordinated actions (e.g. Baldwin 2020, Bénassy-Quéré et al. 2020). Early survey data suggests that uncertainty for businesses has heightened dramatically (Bloom et al. 2020).

In this column, we use new survey data on workers in the US and the UK to show that the impact of the spread of COVID-19 on workers’ current and expected labour supply and earnings is massive.1 However, the impact is not the same for all types of workers. The downturn is particularly harsh for younger workers and those in more precarious types of employment, such as the gig economy. Pressure on people in these segments of society is leveraged by their fear about not being able to cover their bills. The pressure is so high that many report going to work with flu-like symptoms, which poses a health risk to all. We find that this behaviour is particularly likely for those without paid sick leave.

Working from home

Many people are already confined to their homes, some by choice and others by government order. While some workers see minor changes when switching to their home office, for many this means they cannot do their jobs. Figure 1 displays the average percentage of tasks workers report they can do from home by occupation. While in some occupations a high share of tasks can be done from home (e.g. business and financial operations), this is not the case in other occupations (e.g. food preparation and serving). 

Figure 1 Share of tasks that can be done from home by occupation

How are workers in different parts of the income distribution affected by the crisis? As can be seen in Figure 2, individuals with higher annual earnings report being able to do a much higher proportion of their tasks from home. They are also more likely to report that they have already switched to working from home over the past week. 

Figure 2 Share of tasks that can be done from home by annual earnings

Impact on earnings

The ability to work from home can help cushion the impact of the crisis. Whether or not individuals will see a drop in earnings will, however, also depend on their work arrangement and the demand for work. In Figure 3 we see that the self-employed, who typically rely on projects and job assignments, are hit particularly hard. For instance, in the UK, 75% of the self-employed report having earned less last week than in a typical week. In contrast, for salaried workers this number is less than a quarter. Figure 3 only includes individuals who have not lost their jobs. 

Figure 3 Share earning less by work arrangement

While some workers experienced a decline in their earnings but were able to keep their jobs, others report already having lost their job as a result of the crisis. Among all survey respondents who report having had a job four weeks ago, 11.6% and 9.2% report having lost their job due to COVID-19 in the US and the UK, respectively. In Figure 4 we see that job loss has been particularly prevalent amongst workers with low levels of annual income. In the UK, 16% of respondents with incomes below £10,000 report having lost their job due to COVID-19, compared to “only” 5% of workers at the higher ends of the income distribution.  

Figure 4 Share that reports having lost job due to COVID-19 by yearly earnings

Paid sick leave

Before governments decided to officially shut down large parts of economic activity, many decisions were left to the discretion of individual businesses and workers. The decision not to show up at work when feeling ill is particularly difficult for a person without paid sick leave and little savings to fall back on. In Figure 5 we show the share of workers without paid sick leave by occupation.2 We further categorize occupations by the level of physical proximity required in the job using O*NET data. Red bars indicate a high level of physical proximity. We can see that the share of individuals without paid sick leave varies considerably by occupation, and that this share tends to be higher in occupations with a high level of physical proximity. For instance, in the UK more than half of workers in ‘Personal Care and Services’ have no paid sick leave despite working in close proximity to others. Alarmingly, 43% of workers in the UK who do not have paid sick leave report that they usually go to work with a cold or light fever. Moreover, they are 39% more likely to do so than workers with paid sick leave.

Figure 5 Share without paid sick leave by occupation

Outlook for the near future

The evidence we present in this column paints a dire picture of the impact the spread of COVID-19 is already having on workers, in particular for those with unstable employment relationships. Unfortunately, the outlook for the future does not paint a brighter picture. The expected drops in labour earnings, which we show in Figure 6, are striking. The youngest expect to lose nearly half of their labour income in the near future and even amongst those older than 60, who feel the least threatened, the expected drop is almost 30% and 20% in the US and UK, respectively. As a consequence, on average our respondents report that there is a 60% chance they will have difficulties paying their bills in the future.

Figure 6 Median expected income loss in the near future by age

In light of these drastic drops in employment activity and earnings, it seems appropriate for governments to act quickly through stimulus and labour income replacement packages. 


Baldwin, R (2020), “Keeping the lights on: Economic medicine for a medical shock”,, 12 March.

Bénassy-Quéré, A, R Marimon, J Pisani-Ferry, L Reichlin, D Schoenmaker and B Weder di Mauro (2020), “Europe needs a catastrophe relief plan”,, 10 March.

Bloom, N, Bunn, P, Chen, S, Mizen, P, and P Smietanka (2020), “The economic impact of coronavirus on UK businesses: Early evidence from the Decision Maker Panel”,, 27 March.


1  We surveyed 4,003 people in the US and 3,974 people in the UK. Within each country, the samples were selected to be representative of the working population (18+) in terms of region. Given the speed at which events and policy responses unravelled, it is important to situate the moment our surveys were launched. In the US, responses were collected on March 24-25, 2020. At that point, there were more than 55,000 confirmed cases and fewer than 1,000 reported deaths in the US. About half of the US population was already under stay-at-home orders. In the UK, the data were collected on March 25-26, 2020, when there were still fewer than 10,000 confirmed cases and 500 reported deaths in the UK. The lockdown had already been announced a few days earlier, but Prime Minister Boris Johnson had not yet announced the most recent coronavirus relief package. 

2  In the UK, we asked respondents to think about their sick leave entitlement in addition to statutory sick pay. Statutory sick pay of £94.25 per week is available to employees earning more than £118 per week for a maximum of 28 weeks.

26,666 Reads