Has the crisis disrupted thirty-somethings’ career trajectories?

While 82% of the economically active individuals in employment saw no change in their employment situation between March 1 2020 and May 2021, the continuance rate was particularly high (from 86 to 92%) for those who were state employees, on open-ended contracts or self-employed, that is to say those who were already stabilised in their jobs. Most of the self-employed seem to have weathered the health crisis successfully; the level of support available to them undoubtedly contributed to this resilience. Having been established in their positions on average for the same length of time as their employed counterparts, they actually experienced less disruption than the latter group, since 92 % were able to continue their activity.

Nevertheless, the self-employed do not constitute a homogeneous group. The solo self-employed suffered much more from the economic situation, with many more interruptions to their activity. Of the 18% of the economically active population who were employed at the outset and who changed jobs during the period, one quarter stated that they had left their jobs because of the health crisis, whether they had found another job immediately afterwards or had been unemployed for a time. Of these job losses caused by the health crisis, 42% were accounted for by redundancies or terminations of contract and 32% by fixed-term contracts coming to an end. Moreover, for the cohort as a whole, certain experiences such as the postponement or withdrawal of a job offer (15% affected) or the abandonment of a planned business start-up (10 %) cast a cloud over future employment prospects. Only 64% of those who had experienced one or other of these events were actually in employment at the survey date (compared with 86% of those who had experienced neither).

Job tenure and employment status influenced the crisis-related job losses. Employees on temporary contracts were four time more likely than those on open-ended contracts to be affected, and recent arrivals were also more likely to lose their jobs. Nevertheless, longer job tenure did not compensate for the negative influence of precarious employment status (see figure in Box  2). Those individuals obliged to look for work during the pandemic were successful to a greater or lesser extent. Of the economically active without a job on March 2020 and those who had left their job after this date, almost 60% had maintained a certain level of employment up until the survey, having spent at least 7 months in work, with more than 30% of them having spent more than 12 months in work. Conversely, 17% of them experienced severe underemployment, having spent less than 3 months in work during the period. While some - whether they were unemployed (19%) or lost their jobs during the period (17%) - had been able to turn their initial situation to good advantage in order to undergo training, this training (lasting at least two weeks) does not seem to have played a decisive role in their return to employment as observed at the survey date, when 58% of those who had received training were in employment, compared with 69% of those who had not undertaken training.

More than half of the individuals who were looking for work at the beginning of the health crisis (54%), and a higher share of women (57%) than of men (52%), stated that they had not worked at all up until May 2021, while only 18% of them succeeded in working for more than six months during the observation period. In fact, the individuals who had been unemployed in March 2020 were the ones most likely to be unemployed in the spring of 2021. While 34% of the women who were looking for work at the beginning of the period had found a job 14 months later, this share does not seem to be any different from what might have been observed without Covid. And indeed, the data for the 1998 cohort of leavers between March 2007 and May 2008 for a comparable age distribution shows that one third of those who had been unemployed at the beginning of the period were in work 14 months later *.

* This inter-cohort comparison remains relevant even if the economic situation was more favorable for Generation 1998, with a 13% longer duration in employment over 10 years. But, between 7 and 10 years after the end of studies, the situations converge with regard to the cumulative durations spent in unemployment (about 2 months) or the employment rates in permanent job in March 2008 and 2020 (around 87%)

The role of initial educational level in access to employment and the prevention of unemployment is a well-known fact in the early years of individuals’ working lives. During the health crisis, educational level played a particularly important role in enabling individuals to leave unemployment: 64% of higher education graduates returned to employment, compared with just 28% of individuals qualified to baccalauréat level at best. These proportions are reversed for jobseekers who had not worked at all between March 2020 and May 2021: the share of higher education graduates in this group was 24%, compared with 60% of those qualified to baccalauréat level at best. The same calculations applied to the 1998 cohort between March 2007 and May 2008 produced much smaller differences, since 35% of the jobseekers with higher education qualifications returned to work, compared with 31% of those qualified to baccalauréat level at best. At the same time, almost equal shares of the two categories did not work at all during the period. The figures observed for the 1998 cohort show that, while the experience accumulated in the labour market leads to the gradual elimination of the role of initial education in labour mobility, a period of acute crisis such as the Covid pandemic causes tensions to return to the labour market and hence reactivates the inequalities linked to educational level in access to employment.

Those who remained in their jobs between March 2020 and spring 2021 were affected in a different way, since the crisis disrupted the continuity and organisation of their work in particular. Almost a third went on short-time working at some point – a rate more or less equivalent to that observed in the whole of the economically active population in employment (33 % [5]) — and 21 % temporarily stopped work. The factors explaining these events include the contextual characteristics of the sector in question (those of the hospitality industry, for example, other service activities, the distributive trades and automobile repairs, where 50 to 75% of the workforce was put on short time) and company size and location, all of which exerted greater influence than the characteristics linked to employees’ socio-demographic profiles and previous trajectories. These findings suggest that, given the constraints on their activities, companies in these sectors put arrangements in place for the whole of their workforce. The commonality of these employees’ experience, made possible to a large extent by the large-scale support measures for businesses put in place by the French state, constitute a very real difference from the situations of those groups who were without jobs or between jobs during the crisis. Despite everything, the consequences for individuals’ work varied depending on the positions employees occupied within their companies. As in the general population, managerial and professional staff were less affected by short-time working, sick leave and interruptions of activity than the other categories of employee (see table in Box 3). Conversely, managerial and professional staff more often had to agree to imposed leave or reductions in working time (32 %), a compromise that enabled some companies to extend the shorttime working arrangement to other employees. While these mechanisms attest to a reduction in the volume of work, a quarter of employees, conversely, said they had worked more than previously; this applied more to women (30%) than to men (21%) given the sectors in which women are more likely to be employed. This was particularly the case in health and social services (38% of the workforce), a sector under great pressure during the health crisis, and in education (46 %), a finding undoubtedly linked to the need to create new teaching media suited to distance learning.

More than a quarter of the employees had been placed on sick leave (see table in Box  3), a higher share than before the crisis (5.1% in 2019), particularly since in the age groups under consideration absences due to sickness are lower than in the economically active population as a whole  [6]. While women were less affected than men by short-time working (- 5 percentage points) or temporary cessations of activity (- 2 percentage points), they were more likely to be placed on sick leave: at almost 32%, the share was 11 percentage points greater than that for men. This gap cannot be explained by the presence of children in the home, since the sick leave rates were in fact the same for the mothers and fathers of young children, whereas the gap between the sexes was greatest among employees without children (more than 14 percentage points). This disparity is due to differences in the intersectoral distribution of the sexes in employment, since women hold 81% of the jobs in the health and social services sector, a sector where activity was not reduced but where the highest frequency of sick leave among the workforce was recorded (36%). For their part, men account for 85% of jobs in construction and for 70% of the jobs in the manufacturing, extractive and other industries, in which short-time working was more frequent than average (45 and 38% of employees respectively).

From the point of view of work organisation, 38% of the cohort had experienced imposed tele-working. For seven out of ten tele-workers, this represented a new way of working since only a minority had had experience of it prior to the onset of the crisis. Overall, the experience was better for those who had tele-worked before. Unsurprisingly, far more managerial and professional employees were involved in tele-working (four in every five) than the lower occupational categories, whether whitecollar (almost one in every five) or blue-collar (one in 20). Women were more heavily involved than men (41 vs. 34%) because of the kinds of jobs they tend to have. Fathers were more likely to tele-work than men without children: 44% of the fathers of children born before the end of 2017 tele-worked, compared with 32% of those without children. This difference is not observed among women. Thus the family constraints resulting from school closures influenced only male tele-working behaviour. Imposed tele-working was accompanied by an intensification of work for 44% of these employees, equally distributed between the two sexes and 11% percentage points more than for those who had not previously tele-worked. This work intensification could have resulted from a dual trend: on the one hand, an increase in employers’ demands in terms of work intensity and targets, particularly from those employers who mistrusted this mode of working, and, on the other, a splintering of the time actually given over to paid work, particularly during the first lockdown when children were at home all day [7].

One of the aspects that makes this crisis stand out from previous ones is that not all economic activities were affected in the same way by the restrictions imposed during the lockdowns. In this sense, it was the measures imposed on the occupation in question that determined individuals’ risk of being put on short-time or having to stop work altogether rather than personal characteristics such as the level of initial education or previous career trajectory. On the other hand, these latter characteristics continued to influence employee mobility and the ability of the unemployed to find work. The measures taken by the public authorities to encourage tele-working and the various support schemes for businesses and employees did bear fruit and saved jobs. Nevertheless, working patterns were disrupted, with some working more hours, others fewer. Those most seriously affected during the observation period were the ones left on the sidelines by the job preservation schemes, first and foremost job seekers, employees on temporary contracts, the less well qualified and women. However, it seems that children being permanently at home, especially during the first lockdown, had a greater impact on men’s than on women’s tele-working and absences for sickness (in which self-isolation after contact and caring for Covid-positive children have to be included). It is as if the fact of being a parent had changed men’s working patterns, but not women’s. This observation may be explained largely by the need for a more evenly balanced distribution of parental and domestic tasks within couples, particularly when both partners were working in “frontline” occupations.

[1] M. Barhoumi, A. Jonchery, S. Le Minez, P. Lombardo, T. Mainaud, A. Pailhé, C. Pollak, E. Raynaud, A. Sola, « Les inégalités sociales à l’épreuve de la crise sanitaire : un bilan du premier confinement », France Portrait Social, Insee références, Insee, 2020.
[2] Y. Jauneau, J. Vidalenc, « En 2020, après un fort recul lors du premier confinement, le nombre d’heures travaillées s’est plus ou moins redressé selon les professions »,  France Portrait social, Insee Références, Insee, 2021.
[3] M-L. Chausse, M. Gouyon, L. Malard, « Cinq trajectoires sectorielles à l’épreuve de la crise sanitaire en 2020 », Emploi, Chômage, revenus du travail, coll. Insee Références, Insee, 2021.
[4] A. Dupray, I. Recotillet, « Mobilités professionnelles et cycle de vie », Économie et Statistique, n^o 423, 2009.
[5] P. Givord, J. Silhol, « Confinement : des conséquences économiques inégales selon les ménages », Insee Première, n^o 1822, 2020.
[6] C. Inan, « Les absences au travail des salariés pour raisons de santé : un rôle important des conditions de travail », Dares Analyses, n^o 009, 2013.
[7] L. Pelta , « Le télétravail avant, pendant et après la pandémie de Covid-19 », Les Possibles, n^o 27, Attac, 2021.