Alexander Hijzen
Mateo Montenegro
Agnès Puymoyen
Preparing ERTE for the Future
An Evaluation of Job Retention Support in Spain During the COVID-19 Pandemic
Abstract
This chapter provides an overview of the use of job retention support in Spain during the COVID‑19 crisis across different workers, occupations and industries and some of the factors that determine its use.
This chapter provides a detailed picture of the use of ERTE during the COVID‑19 crisis across different firms and workers and the factors that determine its use. The following findings stand out:
ERTE take‑up peaked at 23% of employment in April 2020 and has declined gradually since to 0.12% in December 2022. During most of the period, the bulk of take‑up reflected the full suspension of work. Partial suspensions, however, gradually gained in importance as economic activity resumed.
Small services firms, especially in the hotels and restaurants, as well as the culture sector were the largest users. This stands in contrast to the experience during the global financial crisis when take‑up was concentrated among large manufacturing firms. This most likely reflects the different impact of the COVID‑19 crisis on the ability of different groups of firms to continue to exert their economic activity.
Take‑up was concentrated among workers with permanent contracts at all times but was initially also high among workers with temporary contracts. This is likely to reflect the fact that such workers cannot be dismissed easily before the end of their contract, while incentives to keep temporary workers on ERTE beyond the foreseen end date of their contract are likely to be weak.
Women and initially also young workers were more likely to be on ERTE. The use of ERTE by gender and age is likely to closely reflect the impact of the crisis across sectors and firms and their employment composition across demographic groups. Consequently, the distributional impact of ERTE, at least as far as its use is concerned, is likely to have been modest.
The use of ERTE was stronger in regions, industries and occupations most affected by the COVID‑19 crisis, suggesting support was effectively targeted to firms and workers that needed it most.
Health and economic conditions were important drivers of take‑up. Hospitalisations and deaths due to COVID‑19 were positively related to take‑up, while changes in workplace mobility and economic activity were negatively related. Consequently, ERTE is likely to have contributed to contain the spread of the virus by allowing people to stay at home, while preserving their jobs and supporting their incomes.
Certain jobs that are more exposed to the pandemic such as those involving more face‑to-face contact or jobs that were less teleworkable were associated with a more intensive use of ERTE. However, workers in “essential” jobs, i.e. those (partially) exempt from mobility restrictions, saw a lower use of ERTE, despite being more exposed to the pandemic.
Like in many other OECD countries, job retention support was widely used during the COVID‑19 crisis in Spain. This chapter provides a statistical portrait of the use of job retention support in Spain during the COVID‑19 crisis across different firms and workers and the factors that determine its use. Since detailed information on take‑up is not available for the benchmark countries, the analysis focuses exclusively on Spain. Additional information on the role of programme design for take‑up can be found in Chapter 6 of this report, which provides an in-depth evaluation of ERTE during the COVID‑19 crisis with an emphasis on the role of co-financing, i.e. the cost of reducing working hours under ERTE for firms.
In this chapter, ERTE take‑up will be defined as the ratio of employees on ERTE at any point in time and dependent employment in February 2020 (just before the pandemic hit).
4.2.1. At its peak in April 2020, one in four jobs were supported by ERTE
The use of ERTE was strongly concentrated in the first few months of the pandemic (Figure 4.1). At the start of the pandemic, take‑up shot up immediately, peaking at 23% in April 2020, before declining rapidly to just above 5% in July 2020. In the period from July 2020 to March 2021, take‑up remained broadly stable, at around 5‑7%. From early 2021, it gradually declined to less than 1% by February 2022. Throughout the period, the full suspension of workers has been more common than partial suspension (i.e. reductions in working hours without reaching complete inactivity). However, partial suspensions grew in relative importance in later periods, presumably due to the gradual reopening of the economy, which allowed workers to partially resume their activity. The evolution of take‑up during the COVID‑19 crisis is likely to reflect a combination of health, economic and institutional factors as discussed in more detail below.
Number of jobs supported by ERTE as a percentage of dependent employment in February 2020 by type of suspension, March 2020‑December 2022
Source: Ministry of Inclusion, Social Security and Migration (Ministerio de Inclusión, Seguridad Social y Migraciones).
4.2.2. Take‑up was concentrated in more densely populated regions and important tourist destinations
While the use of ERTE was significant across all of Spain during the first year of the pandemic, take‑up was especially high in provinces with dense populations, notably Madrid and Catalonia, and in important tourist destinations, such as the Balearic Islands, and some regions in the south of the country (Figure 4.2). The use of ERTE declined significantly in all provinces during the second year of the pandemic, but not everywhere to the same extent. For example, top-using provinces in 2020 such as Madrid and Catalonia became average users in 2021.
Take-up in 2002 (March 2020 to December 2020) and 2021 (January 2021 to December 2021) as a percentage of dependent employment in February 2020
Source: Ministry of Inclusion, Social Security and Migration (Ministerio de Inclusión, Seguridad Social y Migraciones).
Smaller firms were more likely to use ERTE
ERTE take‑up was considerably higher in smaller firms, particularly those with less than 50 employees, which represent more than 95% of total firms with dependent employees. One explanation for this pattern is that smaller firms might be more credit-constrained or have less liquidity than large ones. Additionally, as discussed further in Chapter 6, the use of ERTE was initially more expensive for larger firms with more than 50 employees. This coincided with a significant difference in take‑up between firms with 25‑49 employees and those with 50‑99 employees.
The use of ERTE as percentage of dependent employment in February 2020 by firm size, March 2020 to February 2022
Source: Ministry of Inclusion, Social Security and Migration (Ministerio de Inclusión, Seguridad Social y Migraciones).
4.2.3. The use of ERTE was concentrated in the hotels-and-restaurants and the culture sector
The use of ERTE was concentrated in the accommodation and food industry, as well as in the arts, entertainment and recreation sectors (Figure 4.4). In contrast, ERTE was barely used in public administration, household services and the utility sector. The sectoral pattern of take‑up appears to be broadly constant over time. As will be discussed in the next sub-section, these sectoral patterns are driven by the degree to which sectors were affected by the pandemic, either because they could not operate without in-person contact, they could not be adapted to teleworking solutions or they were particularly affected by a reduction in demand.
The use of ERTE as percentage of dependent employment in February 2020 by industry and year
Source: Ministry of Inclusion, Social Security and Migration (Ministerio de Inclusión, Seguridad Social y Migraciones).
4.2.4. Permanent workers, women and initially also youth were more likely to be on ERTE
Who were the workers put on ERTE during the pandemic? As shown in Figure 4.5, take‑up was higher for workers on permanent contracts and women throughout the period. Notably, women were on average 18% more likely to be on ERTE throughout the crisis than men, while temporary workers were 65% less likely to be on ERTE than permanent workers. The pattern for age was more varied. While youth were initially more likely to be placed on ERTE, they were less likely to be put on ERTE from 2021 onwards. One contributing factor could be that access to ERTE and its generosity was progressively restricted, incentivizing firms to put more experienced workers with higher levels of firm-specific human capital on ERTE. The declining importance of ERTE for youth and workers on temporary contracts relative to older workers and workers with permanent contracts is consistent with this explanation.
The use of ERTE as a percentage of dependent employment in February 2020 by group of workers, March 2020 to February 2022
Note: “Young” workers are those aged between 15 and 24, “Prime age” between 25 and 54 and “Older” are between 55 and 75.
Source: Ministry of Inclusion, Social Security and Migration (Ministerio de Inclusión, Seguridad Social y Migraciones).
The nature of the COVID‑19 pandemic as well as the policies enacted to counter it might have generated differences in the demand for ERTE across regions, industries, and time. We study four sets of interrelated factors affecting demand: health shocks, mobility patterns, economic shocks, and occupational characteristics that might make certain workers more vulnerable to the pandemic, such as the likelihood that an occupation involves face‑to-face contact, that it can adapted to telework, or that it was considered essential, and thus was exempted from mobility restrictions during the pandemic.
4.3.1. Higher COVID‑19 hospitalisations and casualties are associated with higher ERTE take‑up
Provinces with higher COVID‑19 hospitalisations or deaths exhibited higher ERTE take‑up (Figure 4.6). In particular, a one‑point increase in the hospitalisation rate of COVID‑19‑patients per ten thousand inhabitants is associated with a 0.27 percentage point increase in ERTE take‑up (Panel A). Similarly, a one‑point increase in the COVID‑19 related death rate per ten thousand inhabitants is associated with a 1.95 percentage point increase in take‑up (Panel B). These positive correlations are likely to capture the possibility that successive waves of the virus generated a demand for ERTE by depressing economic activity. Note that these positive correlations may be muted to some extent by the possibility that ERTE helped to contain the spread of the virus by limiting the need of workers to travel to work and work on-site with others.
The use of ERTE as percentage of employment in February 2020 by autonomous community and month against the number of COVID‑19 related hospitalisations or deaths, March 2020 to December 2021
Note: Each observation is an autonomous community in a given month from March 2020 to December 2021.
Source: National Center for Epidemiology (Centro Nacional de Epidemiología) and Ministry of Inclusion, Social Security and Migration (Ministerio de Inclusión, Seguridad Social y Migraciones).
4.3.2. Reductions in work-related mobility are closely related to ERTE take‑up
Reductions in workplace mobility are negatively related to the use of ERTE (Figure 4.7). Google Mobility Reports can be used to measure reductions in workplace mobility from Q1 2020 to Q1 2022 for each province. This captures the combined effect of administrative restrictions on economic activity and the voluntary mobility responses of workers and firms to minimise the risk of contagion. Relating changes in workplace mobility to the use of ERTE suggests that a 1 percentage point decrease in the workplace mobility index is associated with a 0.56 percentage point increase in ERTE take‑up. As discussed below, one reason why the link between ERTE use and workplace mobility is not stronger still is that some workers switched to telework and hence did not require job support. All in all, this suggests that ERTE contributed to contain the spread of the virus by allowing people to stay at home, while preserving their jobs and supporting their incomes.
Scatter plot of ERTE take‑up (percentage point) against change in workplace mobility index (percentage point) by province and quarter, 2020Q1‑2022Q1
Note: Each observation is an autonomous community in a given quarter from 2020Q1 to 2022Q2.
Source: Google COVID‑19 Community Mobility Reports and Ministry of Inclusion, Social Security and Migration (Ministerio de Inclusión, Seguridad Social y Migraciones).
4.3.3. The reduction in economic activity was a key determinant of the use of ERTE
There is a negative correlation between the change in economic activity between 2019 and 2020 in each industry, measured in terms of value added, on the one hand, and ERTE take‑up on the other (Figure 4.8). As expected, larger economic shocks were associated with a higher use of ERTE. More specifically, a percentage drop in value added is associated with a 0.6 percentage point increase in ERTE take‑up. The pattern is similar to that between workplace mobility and ERTE use and indeed may reflect the same driving factors.
Relationship between ERTE take‑up (2020, percentage point) and the change in gross value added (2019‑20, percentage)
Note: Each observation is an industry. The change in gross value added is from 2019 to 2020.
Source: Ministry of Inclusion, Social Security and Migration (Ministerio de Inclusión, Seguridad Social y Migraciones).
4.3.4. More vulnerable occupations and industries had a higher demand for ERTE
The vulnerability of jobs to the COVID‑19 outbreak is determined, amongst many things, by how much they involve face‑to-face interactions, how much they could be adapted to telework in response to the economic restrictions imposed by the government, and whether they were exempted from such restrictions because they were deemed to be in “essential” sectors that provide basic societal inputs (e.g. food, health) whose provision could not be halted despite the pandemic.1 As seen in Figure 4.9, workers in less teleworkable occupations and requiring more face‑to-face interactions had significantly higher ERTE take‑up, while workers in essential jobs were less likely to be put on ERTE, with sometimes adverse consequences for health and well-being (OECD, 2022[1]).
Note: Each observation is an industry.
Source: Ministry of Inclusion, Social Security and Migration (Ministerio de Inclusión, Seguridad Social y Migraciones).
Since the start of the COVID‑19 pandemic, academics and policy makers have tried to identify which groups of the population are more exposed to its health, income and employment effects. Given the way the virus is transmitted, and the lockdown measures implemented by countries, the degree to which employees can work from home or undertake their normal activities without face‑to-face interactions with others have been key variables identified in the literature. Additionally, belonging to sectors which were considered essential, and thus exempted from mobility restrictions, reduces the short-term negative income and employment effects of the crisis at the expense of running a higher health risk.
In the United States, Dingel and Neiman (2020[2]) and Avdiu and Nayyar (2020[3]) have categorised the degree to which occupations are teleworkable and require face‑to-face interactions, respectively, using survey responses about employees’ daily activities. For instance, occupations in which respondents spend much of their time in outdoor activities, operating large machinery and little time answering emails are given a low score on the teleworkability spectrum, while those that require frequent face‑to-face interactions with the public or assisting and caring for others are given a high face‑to-face interaction score. Although these measures overlap, they are not identical: some occupations might not be teleworkable but do not require face‑to-face interactions, while others are not teleworkable but require personal interactions.
We use occupation-level data with indices on teleworkability and the need for personal interactions from Bossavie et al. (2022[4]) and Garrote Sánchez et al. (2021[5]), who have replicated the strategy for measuring both of these variables for countries in the European Union. We further use occupation-level data on whether certain jobs were considered “essential”, and thus exempted from the mobility restrictions mandated by governments during the pandemic, from Fasani and Mazza (2021[6]). Since no detailed information on take‑up by occupation is available for Spain, we aggregate these indices at the industry level using the composition of occupations by industry in Spain.
This chapter provided a statistical portrait of the use of job retention support in Spain during the COVID‑19 crisis across different firms and workers and the factors that determine its use. The use of ERTE was stronger in regions, industries and occupations most affected by the COVID‑19 crisis, suggesting support was effectively going to firms and workers that needed it most. The take‑up of JR support was particularly high in consumer services (e.g. hotels and restaurants, tourism, culture sector), which require high levels of face‑to-face contact and provide limited scope for teleworking. This also contributed to the relative important of take‑up among small firms, young workers and women.
References
[3] Avdiu, B. and G. Nayyar (2020), “When face-to-face interactions become an occupational hazard: Jobs in the time of COVID-19”, Economics Letters, Vol. 197, https://doi.org/10.1016/j.econlet.2020.109648.
[4] Bossavie, L. et al. (2022), “Do immigrants shield the locals? Exposure to COVID‐related risks in the European Union”, Review of International Economics, pp. 1-32, https://doi.org/10.1111/roie.12609.
[2] Dingel, J. and B. Neiman (2020), “How many jobs can be done at home?”, Journal of Public Economics, Vol. 189, https://doi.org/10.1016/j.jpubeco.2020.104235.
[6] Fasani, F. and J. Mazza (2021), “Immigrant Key Workers: Their Contribution to Europe’s COVID-19 Response”, SSRN Electronic Journal, https://doi.org/10.2139/ssrn.3584941.
[5] Garrote Sanchez, D. et al. (2021), “Who on Earth Can Work from Home?”, World Bank Research Observer, Vol. 36/1, https://doi.org/10.1093/wbro/lkab002.
[1] OECD (2022), The unequal impact of COVID-19: A spotlight on frontline workers, migrants and racial/ethnic minorities.