New research proposes a new classification of occupations based on the extent to which they put workers at risk of being infected by airborne viruses – whether they are ‘safe’ or ‘unsafe’ jobs. Applying the analysis to data from 27 European countries plus the United States, the study by Gaetano Basso, Tito Boeri, Alessandro Caiumi and Marco Paccagnella provides insights into the contrasting health and labour market risks facing different workers.

For example, those who are low educated, self-employed and immigrants are more likely to hold unsafe jobs in essential occupations, and are therefore more at risk of infection. Women and the young are over-represented in the pool of workers in unsafe and non-essential activities, and therefore at higher labour market risk. About 60% of unsafe jobs are in these non-essential occupations: firms restructuring in these sectors may lead to a dramatic drop in labour demand hitting these twice-vulnerable workers.

The analysis has important policy implications. It can be useful in targeting vaccination efforts by occupation, giving priority to workers in unsafe and essential activities, as well as targeting social protection policies for sectors, occupations and firms hardest hit by the pandemic.

More…

Since the outbreak of the pandemic, workers have been facing two types of intertwined risks: an epidemiological risk; and a labour market risk. The former was not given consideration before the pandemic: wages were not compensating for such risk; nor were there objective indicators ranking how safe jobs were along this dimension.

The primary contribution of the new study is to propose a measure of job-related epidemiological risk that unlike other classifications proposed in the emerging body of research on Covid-19 economics, goes beyond the dichotomy between jobs that can or cannot be carried out remotely. In classifying jobs depending on the intrinsic risk of aerial virus infection, the researchers also identify those jobs that, although impossible to be carried out remotely, present a low risk of contagion because they only require infrequent interactions with co-workers and/or customers.

The classification allows meaningful cross-country comparisons. The researchers apply it to 27 European countries and the United States. Safe jobs suitable for home working (overrepresented, for example, among finance professionals) constitute about one-third of all jobs with large cross-country variation. Unsafe occupations, such as medical doctors, waiters and bartenders, represent, on average, 47% of employment.

The remaining share of employment pertains to jobs necessitating some physical proximity and infrequent contacts with other people while still presenting a low risk of contagion. Examples include mixed crop and animal producers, wood processing operators, and drivers of heavy vehicles.

The authors document that their classification predicts well both workers’ actual health and labour market outcomes during the pandemic. Working from home has been concentrated in those jobs that can be potentially carried out remotely, according to the classification.

The ex-ante riskiest jobs experienced the ex-post higher incidence of sick leaves and Covid-19-related injuries. When not associated with ‘essential’ activities (those allowed even during the most restrictive lockdown measures), unsafe jobs were those most involved in subsidised working time reductions (that is, short-time work), and mandatory leave if not outright job loss.

The researchers also document that epidemiological risk is very unevenly distributed across workers and that the pandemic has affected the traditional profile of vulnerable groups in the labour market.

Workers who are low educated, self-employed and immigrants are more likely to hold unsafe jobs in essential occupations, and are therefore more at risk of being infected by the virus. Women, young workers and people in a fixed-term dependent employment position are over-represented in the pool of workers in unsafe and non-essential activities, and therefore at a higher labour market risk. About 60% of unsafe jobs are in these non-essential occupations: firms restructuring in these sectors may lead to a dramatic drop in labour demand hitting these twice-vulnerable workers.

The analysis has important policy implications. It can be useful in targeting vaccination efforts by occupation, giving priority to the workers in unsafe and essential activities, as well as targeting labour market and social policies to the sectors, occupations and firms hardest hit by the pandemic.

It also provides inputs for policies reducing the social costs of adjustment to this new dimension of risk. A mix of policies, such as short-time work and wage insurance, could encourage the mobility of twice-vulnerable workers towards those occupations, and sectors that may offer greater employment opportunities.

On-the-job training and retraining to digital skills can also be very important at the current juncture. Unlike in previous recessions, this time we know which skills are most needed: we should expand the pool of workers who are sufficiently digitalised to be very productive even working remotely.

‘Unsafe Jobs, Labour Market Risk and Social Protection’

Authors:

Gaetano Basso (Bank of Italy)
Tito Boeri (Bocconi University)
Alessandro Caiumi (Bocconi University and University of California, Davis)
Marco Paccagnella (OECD)