Covid-19 hits European cohesion

When the Covid pandemic hit the world in 2020, it seemed for a time that we were all in the same boat. However, job market realities served as a swift reminder that not everyone would be hit equally in Western economies. With specific sectors and some worker categories hit more than others, a number of inequality measures have been at risk of rising ever since. These measures are multidimensional: the Covid crisis and its accompanying episodes of lockdowns have had heterogeneous macroeconomic effects between countries, age and education categories and even between genders. The risk of seeing inequality rise because of the pandemic is real, both in the short and in the longer run.

The risk of seeing inequality rise because of the pandemic is real

Measuring inequality is a long and difficult process: most data is annual (the last Gini indicators date back to 2018 in most cases) or delayed, or lacking the granularity required to make macroeconomic observations between income or age groups in a timely manner. Given the nature of the Covid crisis, we concentrate on country divergences and labour market developments. Eurostat’s labour force surveys, which are available for 2020, allow us to understand what actually happened in the job market and where the largest inequality risks lie, both for the short-term and for the post-pandemic recovery.

During the first months of the Covid crisis, government support measures kept unemployment and employment relatively stable. The eurozone’s contraction in employment in the second quarter (-2.1%) was relatively mild compared to the economic shock (GDP was down 15% compared to 4Q19) or what was observed in the US (-12.8% of total employment). This is because workers on temporary unemployment schemes (“furloughed” workers) were actually not counted as unemployed: they were unable to go to work, sometimes losing a sizeable share of their incomes (as benefits didn't always cover all of their revenue losses), but at the same time they were not looking for a job, as they still had one. This large population - representing 32 million workers at the peak or three times the number of unemployed at the same time – was not counted in the unemployment statistic.

To get a grasp of what happened in the labour market, we look at Eurostat’s labour force surveys. They show that hours worked actually diverged significantly across eurozone countries. In the eurozone as a whole, hours worked dropped by 17% in 2Q20 (compared to 4Q19), but the decline ranged from 7.5% in the Netherlands to 26% in Greece. Figure 1 also shows that while the shock had a very different impact between countries, the third quarter put countries back on a similar footing. However, if divergences are less obvious in 3Q20, they nevertheless remain relatively high: hours worked in 3Q20 ranged from 97% of pre-crisis level in Belgium and the Netherlands to only 93.2% in Spain, for example.

This data shows that the effort needed to catch up to pre-crisis levels is still very high: during the financial crisis, it took 10 quarters (from mid-2013 to the end of 2015) to return from 95% to 100% of pre-crisis hours worked in the eurozone. To be sure, part of this slack is directly linked to lockdown measures and once these measures have disappeared with the vaccination campaign, some catch-up will occur. However, we believe that it's very likely that the labour market slack, measured in hours worked, is concentrated in parts of the labour market (see below), which – given the lengthening of the lockdown situation in 2021 – could end up having a long-term impact on inequality.

The Covid-related lockdowns had an economic impact that was heavily concentrated in sectors where demand relies on mobility or human contacts (manufacturing of vehicles and transport material; wholesale and retail trade; hotels, restaurants and air travel; professional and real estate services; arts and entertainment). These activities represent up to 40% of the national gross value added in Greece, and less than 25% in Ireland.

We find that the divergence in the hours worked contraction is strongly related to countries' specialisation in these vulnerable sectors: for example, they make up to 40% of economic activity in Greece, Spain and Portugal, partly explaining why these countries saw hours worked plummet by 25% in the first weeks of lockdown. We also note that the relationship holds for 3Q20.

It appears that these present hard-hit sectors share characteristics across countries which make them vulnerable to a rise in inequality: these sectors are indeed intensive users of non-standard contracts (where low-educated and young workers are concentrated) and low-paid jobs. The former represents 30.5% of employment in Covid-vulnerable sectors in the eurozone, the latter 23%, compared to 26.8% and 15%, respectively, in total employment (all sectors aggregated).

We find, for example, that the gaps are larger in Italy, the Netherlands, Portugal and Belgium where the vulnerable sectors make much larger use of NSW contracts than the average, with proportions reaching 40% in the Netherlands and Spain and 43% in Italy. When it comes to low-paid jobs, other countries stand out: the share of low-paying jobs in vulnerable sectors is disproportionately high (compared to the national average) in Austria, Germany and Ireland, with 36% of employment in vulnerable sectors being low-paid jobs in Germany, against a eurozone average of 23%.

Vulnerabilities can be distributed differently from one country to another

Given that countries have different degrees of specialisation in vulnerable activities, the asymmetric shock on worker groups also implies divergence in inequality risk among countries. We show the risks of seeing a surge in inequality due to Covid in two dimensions along which we can map countries: the share of employment covered by either non-standard or low-wage contracts in vulnerable sectors.

This shows that vulnerabilities can be distributed differently from one country to another: low-wage workers in Germany and Ireland, non-standard contract workers in Portugal and Spain, and both in Greece and the Netherlands.

The growth patterns observed in the eurozone economy in the second half of 2020 show that the recovery, so far, has been as imbalanced as the shock itself. Some sectors have seen their gross value added (GVA) catch up to and even sometimes surpass their pre-crisis level, while the output gap remains concentrated in the most vulnerable sectors. Figure 4 confirms that in most countries, vulnerable sectors are lagging the rebound that occurred in the rest of the economy: in Spain, vulnerable sectors still have a GVA that is 15% below pre-pandemic levels while the rest of the economy is only 2% below that level. In Germany, Portugal, the Netherlands and Ireland, the rest of the economy has even fully recovered.

This is important because it means that the “90% economy” highlighted by The Economist one year ago does not exist: most sectors have (almost) recovered fully, while some of them are living in an “85% economy”. In terms of inequality, it heightens the risk as these sectors are heavy users of more vulnerable forms of employment. What is more, we know that given the slowness of the vaccination campaign in Europe, lockdown measures will take time to disappear so the crisis will last longer for these sectors. While government measures taken in the first few months of the pandemic may have worked for all sectors in a time-limited shock, it is likely that the 18-month shock faced by vulnerable sectors will require more specific measures, if the risk of rising inequality is to be contained.

There are several reasons why we think a greater part of the working population is at risk of falling into poverty as long-term inequalities persist.

• First, lockdown measures have been renewed throughout the continent recently, hitting the same sectors again in the first half of 2021.
• Second, there is no prospect of getting back to levels of human contact that allow these sectors to work at capacity before the end of 2021 when most of the population will be vaccinated.
• Third, post-pandemic times could potentially bring a toxic “new normal” for workers in vulnerable sectors as the number of employers will have shrunk through bankruptcies, probably capping wages and contract durations for longer. Low-wage and NSW workers in vulnerable sectors represent 5% to 15% of employment in European countries, which is far from negligible.

To counter these risks, some countries have taken measures to ensure that vulnerable workers who have fallen through the safety net do not fall into poverty. But some countries could do more. What our findings show is that after the broad-based measures, which were justified at the beginning of the pandemic, more targeted measures aimed at specific groups of workers now need to be planned until year-end. As it is much easier to fall into poverty than to get out of it, forthcoming recovery plans will have to focus on vulnerable employment, which has not been addressed by other measures.