Comment expliquer l’évolution de l’emploi salarié depuis la crise Covid ?. Une analyse économétrique sur données macro-sectorielles

Since the outbreak of the pandemic, the dynamism of job creation in the non-agricultural market sector in France has continued to surprise us quarter after quarter. At the end of 2022, three years after the start of the Covid crisis, activity in this sector, measured by its value added, was 1.2% above its pre-crisis level. Given the growth path of labour productivity observed before the crisis, estimated at 0.9% per year by Ducoudré and Heyer (2017), this weak growth in activity should have led, all else being equal, to a 1.6% fall in paid employment in the market sector by the end of 2022. But instead of falling by more than 270,000 jobs, non-agricultural paid employment in the market sector has actually risen by almost 800,000 (+4.6%) over the last three years, according to national accounts data. In this article, three possibilities have been put forward and tested to explain this gap of more than 1 million jobs in the market over the period 2019-2022: ? The first is that average working time per employee has still not returned to its pre-crisis level, reducing the apparent productivity of employees; ? The high take-up of apprenticeships observed since 2019 is a second possible explanation (Coquet, 2023); ? The third could be the numerous subsidies distributed to companies since the Covid-19 crisis which, by changing the incentives for companies to lay off and hire, may have encouraged them to retain labour. Not only could this aid have been able to boost employment growth in companies that are doing well, but it could also artificially have kept some of them in business even though they should have gone bankrupt, as illustrated by the very low number of company failures over the last three years. Based on our estimates of labour demand equations using macro-sector data, these three factors explain almost 70% of the difference in job creation described above. Breaking this down, the shorter working hours of employees explain 18% (i.e. almost 200,000 jobs), 24% is linked to the sharp rise in the number of apprentices over the analysis period (i.e. more than 250,000 jobs) and 26%, i.e. almost 280,000 jobs, is due to the ?exceptional? business support measures. It should also be noted that while these three factors explain almost all of the difference in the market services sector, they explain only half of it in the construction sector and barely a third in industry. In this latter sector, before concluding that there is a downward trend in productivity gains and while awaiting the definitive national accounts, an alternative explanation could be that the anticipation of an upturn illustrated by full order books could be encouraging employers to retain their workforce in order to avoid the costs associated with finding new candidates once supply problems have been resolved. This behaviour may be heightened in a context where a large majority of companies are reporting labour shortages.