https://orcid.org/0000-0002-1660-5165
ABSTRACT
This article discusses the evolution of key labour market indicators in the EU-27 countries between 2019 and 2020, i.e. between the year before the covid-19 crisis broke out and the year in which it impacted the economy heavily. Whereas earlier studies have dealt with the evolution of unemployment in 2020, often country by country, this article focuses on describing the evolution of unemployment as well as inactivity across European countries. It appears that the Southern European countries, in particular, recorded increases in inactivity, while the Baltic States experienced higher growth in unemployment. In many other countries, unemployment and inactivity remained remarkably stable despite covid-19. The country comparison bears many similarities to that in 2009, when the Great Recession knocked back the European economy.
I. Introduction
The year 2020 will undoubtedly be remembered by many as the ‘covid-19 year’. After all, the impact of this pandemic on the lives of citizens was enormous, as was the economic shock. Rarely has the economy of the OECD countries contracted as much in a single year as it did in 2020. In the EU-27, the average drop in real GDP was 6.1% (source: Eurostat, ‘Real GDP growth rate – volume’).
Based on previous crises, we know that the labour market typically ‘follows’ GDP patterns. That is, a decline in the demand for goods and services within the economy translates into a decline in demand for labour (Kamar, Bakardzhieva, and Goaied Citation2019). However, the latter decline is often delayed, as firms initially try to avoid restructuring and layoffs through so-called ‘labour hoarding’ (Biddle Citation2014). It is therefore uncertain as to whether the economic calamity brought about by the covid-19 crisis translated into labour market havoc in 2020.
The comparison of the performances of labour markets over time and between countries is often based on unemployment rates. Thus far, the short-term effects of the covid-19 crisis have been mostly traced through the evolution of this indicator over the months of 2020 and early 2021 (OECD, Citation2020). However, as argued in Baert (Citation2021), doing so may yield a biased comparison. Indeed, the unemployment rate, by definition, indicates the percentage of the active population (i.e. employed or seeking work) without jobs at a given moment. In other words, those who neither have a job nor are looking for one (the ‘inactive’) do not appear in the numerator or the denominator of this calculation. As a result, two countries may have the same unemployment rate but different employment rates (i.e. the percentage of employed persons out of the total population within certain age categories). In Baert (Citation2021), we compare this situation to an iceberg: for decades, policy has focussed on the visible labour reserve represented by the unemployed but forgotten about the latent reserve of inactive people below the water line. The size of this group of inactive people, however, has major implications for public financing, as inactive people typically do not contribute to it but are often supported by it.Footnote1,Footnote2
Moreover, during a crisis period, unemployment and inactivity follow different dynamics in terms of their development. As indicated earlier, rising unemployment is the direct consequence of a fall in the demand for labour. In contrast, rising inactivity during a crisis is linked to discouragement. A limited increase in unemployment during the first months of the covid-19 crisis could therefore be hiding the fact that a segment of the unemployed simply gave up looking for a job.
In this article we therefore describe how EU countries saw their labour markets evolve in 2020 according to the percentage of unemployed among the entire group of 25- to 64-year-olds (‘unemployment-to-population ratio’) and the corresponding percentage of inactive persons (‘inactivity-to-population ratio’). We will zoom in on how the rankings of European countries for these two indicators changed between 2019 and 2020. To the best of our knowledge, we are the first to describe these evolutions of both measures jointly. Moreover, we contribute to the peer-reviewed economics literature by comparing these evolutions with those in 2009, when the European economy experienced a negative GDP shock in the context of the Great Recession.
II. Data
The analyses in this article are based on the figures that Eurostat publishes each year regarding: (i) the percentage of employed persons within various age categories, (ii) the percentage of inactive persons within various age categories, and (iii) the main reasons provided by these inactive persons for their inactivity. Regarding (iii), more specifically, we retained the percentage of individuals with the ‘belief that jobs are not available’. For each of these statistics, the 25- to 64-year-old age group was examined. These data were released by Eurostat on 21 April 2021. Appendix summarizes these source data.
III. Results
The iceberg in 2020
analyses the proportions of unemployed and inactive people in 2020 according to the so-called ‘iceberg decomposition’, i.e. the entire population aged between 25 and 64 is divided into three groups: the employed, unemployed and inactive people. The unemployment-to-population ratio is calculated on the basis of the aforementioned source data by subtracting the percentages of employed and inactive people from 100%.
Figure 1. Iceberg decomposition for 2020.
At the EU-27 level, the proportion of inactive persons (i.e. 20.3%) is more than four times higher than the proportion of unemployed (i.e. 5.0%). The lowest percentage of unemployed is 1.9% (in the Czech Republic) and the highest is 11.8% (in Greece). The percentage of inactive persons varies between 10.8% (Sweden) and 28.6% (Italy).
Evolution of the iceberg between 2019 and 2020
Of course, what is of interest is the evolution of these fractions between 2019 and 2020. In line with the previously mentioned reports of rather limited increases in unemployment during 2020, shows an increase in unemployment in the EU-27 of only 0.2 percentage point (pp; from 4.8% to 5.0%), on average.
Table 1. Evolution of the unemployment-to-population ratio (among 25- to 64-year-olds).
This average obviously hides differences between the EU countries. Most strikingly, in the Baltic States, the increase is more than 1.5 pp and therefore substantial: Estonia (1.7 pp), Latvia (1.8 pp) and Lithuania (1.7 pp). Besides, only Romania (1.0 pp) and Sweden (1.2 pp) exhibit a growth in their unemployment-to-population ratios of 1 pp or more. These countries also drop down in the ranking of countries according to this ratio. For instance, Estonia drops seven places (from position 11 to position 18). Sweden even ends up in the worst quartile in 2020 (unemployment-to-population ratio of 5.9% in 2020; ranking: 23 out of 27). Countries moving up in the ranking include Belgium (from 14th to 10th position), France (from 24th to 20th position) and Slovenia (from 12th to 8th position).
Although an increase of a few tenths of a pp in the unemployment-to-population ratio typically implies many thousands of additional unemployed, these changes can be considered limited. By comparison, in Great Recession year 2009, when the real GDP growth in the EU-27 was −4.3% (while it was positive in 2008 and 2010), the unemployment-to-population ratio increased by 1.3 pp at the EU-27 level.
What happened to the percentage of inactive people in 2020? Overall, shows that the increase in the percentage of inactive persons at the EU-27 level also remained rather limited. In 2019, 20.0% of the population was considered inactive; in 2020, as indicated earlier, that percentage rose to 20.3%. In other words, there was an increase of 0.3 pp. This increase does imply, as Appendix Table A.1 indicates, that the downward movement in the percentage of the population considered inactive since 2017 has been reversed. However, in absolute numbers, a 0.3 pp increase in inactive persons in the 25- to 64-year-old group is of course substantial: an increase of about 720,000 inactive persons.
Table 2. Evolution of the inactivity-to-population ratio (among 25- to 64-year-olds).
Again, however, we see important differences between countries. Inactivity rose more sharply in Southern Europe: Spain (1.1 pp), Italy (1.5 pp), Portugal (0.6 pp) and Greece (1.0 pp). Bulgaria (0.8 pp) and Ireland (0.8 pp) are also close to 1.0 pp increases in inactive persons. Interestingly, the Baltic States, with, as mentioned previously, the largest increases in their unemployment-to-population ratios, all did favourably in terms of the percentage of inactive persons, which even fell by 1.4 pp in Latvia.
summarizes the movements in unemployment and inactivity of the 12 largest EU countries, i.e. those countries with populations over 12 million (source: Eurostat, ‘Population on 1 January’). A number of country clusters can be distinguished. First, in Belgium and the Netherlands, unemployment and inactivity remained largely at the same level between 2019 and 2020. Second, there are countries for which unemployment rose but inactivity remained at the same (low) level: the Czech Republic, Germany and Sweden. Third, in Spain and Portugal, both unemployment and inactivity rose, albeit more sharply in Spain than in Portugal. Fourth, two other southern European countries also saw an increase in inactivity, but it was accompanied by a decrease in unemployment: Greece and Italy. The same is true, albeit to a lesser extent, for France. In Romania, the opposite happened: higher unemployment but (slightly) lower inactivity. Finally, Poland is a ‘unique case’ too – but a favourable one – inactivity and unemployment slightly decreased.
Figure 2. Evolution of the unemployment-to-population and inactivity-to-population ratios (among 25- to 64-year-olds) in the 12 largest EU countries.
Do these shifts go in the same direction as those in 2009 (Great Recession; see above)? In other words, are the same countries as then recording higher increases in unemployment and inactivity? Table A.2 and Table A.3 show the corresponding measures for 2008 and 2009. The similarities in evolution are striking, especially in terms of the unemployment-to-population ratio. Also in 2009, the top-3 in increases in this measure were made up by the Baltic States (between 6.0 and 7.1 pp higher) and the increase in Sweden was higher than the EU-27 average. The correlation between the evolution in 2020 on the one hand and that in 2009 on the other is high: Pearson’s r is 0.661. With respect to the evolution in inactivity-to-population ratio, it is 0.187. Remarkably, as far as this second measure is concerned, it are again Bulgaria and Ireland that have a clear increase in 2009; as far as the Southern European countries are concerned, the picture in terms of evolution of inactivity is more diffuse. We return to this point when we formulate policy reflections in Section 4.
Importance of the discouraged unemployed among inactive persons
Basically, in terms of the fraction of inactive people, the covid-19 crisis did not hit most countries substantially in 2020. This is a first indication that the discouragement among the unemployed has not been that bad. A second indication is provided in . This figure shows the percentages of individuals 25–64 years old that were inactive due to the belief that there would be no jobs, for the 12 largest EU countries in 2019 and 2020.
Figure 3. Changes in the percentage of 25- to 64-year-old inactive persons with as a main reason the belief that jobs are not available for the 12 largest EU countries.
makes it clear that the fluctuations in the percentages of discouraged people are very limited. The largest increase is found in Greece, which went from 0.6% to 1.0%. On the level of the entire EU-27 (Appendix Table A.1), this percentage remained stable at 1.1%.
IV. Conclusion
The figures in this article indicate that, in terms of unemployment-to-population and inactivity-to-population ratios, most European countries did not receive a huge blow from covid-19 in 2020. Still, in absolute numbers of citizens, the increase of 0.3 pp in the percentage of inactive persons implies an increase of about 720,000 persons. Moreover, there are important differences between countries: inactivity rose more sharply in Southern Europe, while unemployment rose more sharply in the Baltic States.
It is further noteworthy that there is a clear positive correlation between the evolution of both measures in 2020 and that in 2009, when in the EU-27 real GDP growth took a dive of 4.3% in the context of the Great Recession. This indicates that some EU states, the Baltic States in particular, are a lot more shock-prone than others and can learn from countries that digested the shock well in 2020. It is also important to note that there does not seem to be a trade-off between shock resilience and structurally low unemployment and inactivity. The correlation between the evolution in unemployment-to-population ratio in 2020 and the level of this measure in 2019 is clearly positive (Pearson’s r = 0.514).
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 We also explain this logic and the various indicators for measuring the health of a labour in an animated video that can be viewed at www.stijnbaert.eu or directly via this YouTube-link https://youtu.be/DDICIu5ykrc
2 During the last decades, the employment rate has become more prominent as a crucial indicator of labour market performance – as indicated by an anonymous reviewer of a former version of this article, it has been the main indicator suggested by the European Employment Strategy launched in 1997 (Council of the European Union Citation1997). However, this indicator is still somewhat less prominent in debates during a crisis period given that dismissals obviously translate into unemployment – as an illustration, ‘employment rate’ in combination with ‘covid-19’ yields 16.5 million Google hits on 14 July 2021, while for a combination with ‘unemployment rate’ this is more than 20.3 million. Moreover, Baert (Citation2021) argues that also the employment rate is not an indicator that adequately captures the health of a labour market since job seekers and inactives are lumped together. For an overview of other relevant indicators, we refer to Brandolini and Viviano (Citation2018) and ILO (Citation2016).
References
- Baert, S. 2021. “The Iceberg Decomposition: A Parsimonious Way to Map the Health of Labour Markets.” Economic Analysis and Policy 69: 350–365. doi:https://doi.org/10.1016/j.eap.2020.12.012.
- Biddle, J. E. 2014. “Retrospectives: The Cyclical Behavior of Labor Productivity and the Emergence of the Labor Hoarding Concept.” Journal of Economic Perspectives 28 (2): 197–212. doi:https://doi.org/10.1257/jep.28.2.197.
- Brandolini, A., and E. Viviano. 2018. “Measuring Employment and Unemployment.” IZA World of Labor 445.
- Council of the European Union. 1997. The 1998 Employment Guidelines. Brussels: Council of the European Union.
- ILO. 2016. Key Indicators of the Labour Market. Ninth ed. Geneva: International Labour Office.
- Kamar, B., D. Bakardzhieva, and M. Goaied. 2019. “Effects of Pro-growth Policies on Employment: Evidence of Regional Disparities.” Applied Economics 51 (40): 4337–4367. doi:https://doi.org/10.1080/00036846.2019.1591596.
- OECD. 2020. “Worker Security and the Covid-19 Crisis.” In OECD Employment Outlook 2020. Paris: OECD.
Appendix
Table A1. Source data
Table A2. Evolution of the unemployment-to-population ratio (among 25- to 64- year-olds) during the Great Recession
Table A3. Evolution of the inactivity-to-population ratio (among 25- to 64-year- olds) during the Great Recession
![Supercharge Your Next Research Paper: Research and write your next paper with Jenni AI.]({"type":"image" , "src" : "/sda/112446/MPU-L1EB.png"})