VoxEU Column COVID-19 Labour Markets

The asymmetric impact of COVID-19 confinement measures on EU labour markets

The COVID-19 economic crisis is having a huge impact on employment in the EU, calling for swift policy action targeting the most affected sectors and countries. This column makes an assessment of the labour market impact of the confinement measures put in place by EU governments. It finds that these restrictions are likely to have a very asymmetric effect across EU labour markets, with the most negative employment impact concentrating in the most vulnerable countries and categories of workers.

The COVID-19 economic crisis is an unprecedented shock to European economy and society that requires swift policy action and a coordinated fiscal response (Baldwin and Weder di Mauro 2020). The EU is currently designing a package of measures to mitigate the impact of the COVID crisis. A crucial element of the policy mix is to mobilise investments through the so-called recovery fund, which EU leaders agreed upon at the end of April and is now being designed to target the most affected sectors and geographical parts of Europe.

This crisis is already having a huge labour market impact. As a result of the economic shock and the measures implemented to halt the spread of the virus, millions of workers have lost their jobs or have been placed under short-term work schemes. Millions more are teleworking, with emerging challenges in terms of work-life balance and working conditions. A good understanding of the different labour market implications by countries, sectors, and groups of workers is crucial to ensure that policy action reaches those who are most in need of urgent support.

In this column we contribute to this by summarising the findings of a recent study on the potential impact of the early 2020 confinement measures on EU labour markets (Fana et al. 2020). The analysis is based on an assessment of the restrictions on activities imposed in three EU member states (Italy, Spain, and Germany), which are used as benchmarks for the rest of Europe. Following the legislative measures adopted in those three countries as first response to the spread of the virus, all economic sectors are classified into different categories according to the likely impact of the COVID-19 crisis, and then we compare the shares of employment that are likely to be affected in each European country. On this basis we estimate how different types of workers are likely to be affected by the economic lockdown measures and discuss the potential socio-economic implications of the COVID-19 crisis in Europe.

The unequal effect of the confinement measures

In order to assess the impact of the crisis by sector and type of worker we base our analysis on five categories of sectors according to the likely impact of the confinement measures (see Annex for details on the construction of this classification). These categories of sectors are:

1) essential and fully active;

2) active but via telework;

3) mostly essential and partly active, but not teleworkable;

4) mostly non-essential and inactive, not teleworkable; and 

5) closed.

This classification was applied to an ad-hoc extraction of the EU Labour Force Survey data (with annual figures from 2018), in order to compare the share of employment potentially affected in different countries (figure 1).

Figure 1 The distribution of employment across the 5 categories of sectors, defined by likely impact of COVID crisis.

Source: Employment figures from 2018 annual LFS data

The countries with a higher share of employment in the forcefully closed sectors are likely to suffer a much higher impact. As we see in Figure 1, for the EU as a whole, the forcefully closed sectors represent around 10% of employment, but there are significant differences by country, with some Southern European countries plus Ireland showing the highest shares. The main driver behind this heterogeneity is regional economic specialisation. Indeed, some Mediterranean countries account for higher shares of employment in leisure activities, hospitality, personal services, and other sectors that have been strongly hit, but they also have higher shares of self-employment and temporary contracts (especially in the closed sectors), which can compound the negative effects of forceful closures. On the other hand, the countries in which the shares of essential or teleworkable sectors are higher are those of Northern and Western Europe. These countries are potentially less exposed to the negative consequences of the current crisis.

An analysis of the socio-economic composition of the groups of sectors defined above allows us to deepen on the distributional consequences of the COVID crisis. Our study suggests that the effects of the lockdown measures across groups of workers are markedly asymmetric. In short, the most negative effects of the confinement measures are often concentrated on the most vulnerable and disadvantaged workers. This is because the sectors forcefully closed (5) by the decrees (i.e. hospitality, personal services, leisure activities, etc.) are, in most EU countries, characterised by low wages and precarious conditions of employment, and tend to have a higher concentration of women and young workers than the rest. For illustration, Table 1 shows the average wage percentile of jobs in the different categories of sectors in Europe.

Table 1 Average wage percentile of jobs in each of the categories

Source: Employment figures from 2018 annual LFS data. Wage percentiles by job from EJM database (Hurley et al. 2019).

In contrast, there is a category of sectors for which the restrictions imposed by confinement measures are not so adverse. This category is formed generally by service sectors that involve some degree of social interaction (paradigmatic examples are education, public administration, telecommunications, and most professional, scientific and technical activities) that lends itself to remote service provision. In sharp contrast with the forcefully closed activities, these ‘teleworkable’ sectors (2) are characterised in most countries by better than average employment conditions: the proportions of self-employed and temporary workers are comparatively lower, and they have very high wages (see Table 1). A remarkable share of high-skilled workers is active in this category of sectors.

In between those most and least negative outcomes, there are three more categories of sectors defined according to the COVID-19 lockdown measures. The characteristics of those sectors are summarised below.

Sectors considered essential and thus remaining fully active (1): This category includes activities related with food production, utilities, the activities of health professionals and social care, etc. They account, on average, for around 25% of employment in most countries, with conditions and wage levels generally similar to those of the average working population. In this case, there is an important gap by age: the category of sectors that was fully active and open to the public even when most restrictive measures were applying is the one employing the highest proportion of older workers, while young workers are underrepresented.

A mixed category of sectors which are partly considered essential and thus partly active (3): Here we include most of retail and manufacturing of chemicals and paper, which remain to some extent active even in a strict confinement situation. The typical conditions of this category are similar to those of the forcefully closed sectors, and in some cases may suffer similar problems in the short and medium term. For instance, non-essential retail is also likely to suffer forceful closures and intermittent activity in the near future.

The activities that are not considered essential but, in many cases, are allowed to continue functioning with additional precautionary measures (4): Here we include activities that usually do not involve high risks to the general public but which cannot operate on remote since they require the physical transformation of objects, including the majority of manufacturing not previously mentioned, some machine and computer repair activities, and construction. These typically male-dominated sectors tend to have better employment and wage conditions than the average, even if their average education levels are in fact below those of the forcefully closed sectors. Although these sectors may also suffer a significant blow because of the economic contraction, the lockdown measures themselves are unlikely to affect them in a significant way in the medium term.

Labour market impacts and policy implications

Our analysis suggests that the impact of the COVID-19 crisis is likely to be concentrated in some countries and, generally, on the most vulnerable segments of the working population. Restrictions on economic activity are mainly affecting workers with lower wages and worse employment conditions. The impact also appears to be more significant for women and young workers. It is important to notice that these segments of the working population are also probably the ones with less resources available to face unemployment and sudden income losses. The challenges for unemployed people are likely to be quite significant in the short and medium term, since they will have to look for jobs in a context of subdued economic activity and employment scarcity.

A collective EU response based on pan-European emergency mechanisms is being set up to provide support for the countries most in need. In the mid-term, the economic sectors most affected now will remain problematic until the pandemic is under control, because they involve an important degree of face-to-face social interaction and final (often external) demand. A very significant proportion of the workers now employed in those sectors may face very uncertain prospects in the medium term. The policy response should therefore not be limited to the short term, and should include investment policies that provide alternative opportunities at a large scale, oriented towards long-term strategic goals such as an ambitious European Green Deal.

Authors' note: The views expressed in this column are those of the authors and cannot be taken as representing the official position of the European Commission.

Annex: Classification of sectors according to the confinement decrees in Spain, Italy and Germany

For details, please see Fana et al (2020). Table 2 presents the analysis of the containment decrees by sector in Germany, Spain and Italy. For each sector and country, we provide an indicator that ranges from 0 to 1 (columns DE, ES and IT). These values have been qualitatively assigned by careful analysis of the decrees: in the majority of cases, the decrees explicitly mention the sector, although in some cases we had to infer the value. A value of 1 indicates that the sector is explicitly defined as essential, and thus can continue to operate even in the strictest confinement. A value of 0 indicates that the sector is considered non-essential, which may mean that it is forcefully closed or that it can operate only under certain conditions (more on this later). Values between 0 and 1 indicate that some sub-sectors (NACE 3 or even 4 digit codes) within a given sector (NACE at 2 digits, which is our baseline) are considered essential and some not: in these cases, the value of the indicator reflects the share of sub-sectors considered essential, when possible adjusted for relative employment shares.[1] The column "Index" is a simple arithmetic average of the three country columns, and thus can be interpreted as the average degree to which a given sector is considered essential in the three countries analysed.

As can be seen in Table 2, this average indicator has been used to rank the sectors, providing a first criterion to classify them according to the impact of the COVID confinement decrees. But to make a better assessment and classification, we need to consider two additional criteria which also derive from our analysis of the confinement decrees.

First, whether a given sector can operate via telework. This mostly depends on the nature of economic activity in each sector: in general, activities and services that do not involve direct physical interaction (either with things or with people) can be remotely provided making use of ICT equipment. All the confinement decrees state explicitly that independently of whether a given sector is considered essential or not, whenever possible it should operate via telework. For instance, the education sector is considered essential in two of the three countries analysed, yet in all three cases schools are forcefully closed and educational activity continues via the internet.  There are also sectors such as advertising and market research that are considered as non-essential in all three countries yet in general can continue operating because physical presence is not critical. In practice, what this means is that irrespective of their classification as essential or not, all teleworkable sectors can continue in operation. For this reason, as indicated in the column "Notes" of Table 2, we have added a category of "teleworkable" (code 2) that spans throughout the whole table. We have included in this category education and public administration, as well as most professional services (ICT, insurance, financial, accounting, programming, research, design, advertising).[2]

Second, although the main distinction in the analysed decrees is between essential and non-essential activities, in practice there is also an implicit or explicit differentiation between those (non-essential) activities that are forcefully closed because they require direct face-to-face interaction with clients and thus they are particularly risky in the context of the COVID pandemic. These explicitly and forcefully closed activities include hospitality, accommodation, leisure and personal services. It is important to differentiate these activities from those that are also non-essential in the three countries but which do not involve a particularly high risk of contagion (for instance, most of manufacturing and construction) and thus can in most cases continue to operate even if under strict conditions. Therefore, in table 2 the activities which are fully or mostly non-essential (values below .3 in the indicator) are classified in two different categories: those that are forcefully closed (5), and those that are mostly non-essential but not forcefully closed (and thus at least partly active, code 4). The distinction between the latter two categories does not apply in situations of very strict confinement as in Spain between 30th March and 9th April, when all non-essential sectors were forcefully closed.

Table 2 A summary of the COVID sector lockdowns in three European countries as of early April 2020

Key: 1 means essential in the respective national decree; 0 non-essential, and fractions the share considered essential in each sector. The index is a simple average of the three values by country.


Baldwin, R and Weder di Mauro, B (eds.) (2020), Mitigating the COVID Economic Crisis: Act Fast and Do Whatever It Takes, CEPR Press.

Fana, M, Tolan, S, Torrejón, S, Urzi Brancati, C and E Fernández-Macías (2020), “The COVID confinement measures and EU labour markets”, EUR 30190 EN, Publications Office of the European Union, Luxembourg. ISBN 978-92-79-18366-2, doi:10.2760/69199, JRC120578.

Hurley, J, Fernández-Macías, E, Bisello, M, Vacas, C and M Fana (2019), European Jobs Monitor 2019: Shifts in the employment structure at regional level, Eurofound and European Commission Joint Research Centre, Publications Office of the European Union, Luxembourg. 


1 For instance, in Spain sector 19 (manufacture of coke and petroleum) has two subsectors: one which is considered essential (manufacture of refined petroleum) and one which is not (manufacture of coke). Since the relative weight (in terms of employment and economic activity) of coke manufacturing is much smaller in Spain than that of refined petroleum, we assigned a value of .9 to sector 19 in Spain.

2 In a forthcoming paper, we analyse in detail teleworkability by occupations in Europe, using a tasks approach.

5,669 Reads