Job Loss in a Group of Older Canadian Workers : Challenges in the Sustainable Labour Market Reintegration Process

In Western countries, the loss of jobs among older workers is a highly worrisome situation, since it can be synonymous with long-term employment precariousness and definitive exclusion from the labour market. This precariousness is occurring while the labour force in these countries is aging, and governments are looking to extend people’s working lives. It is therefore particularly relevant to study different labour market reintegration processes and to understand their sustainability from a psychological perspective. The present article is examining these processes using a longitudinal study over an 18-month period with 61 older Canadian workers. Time 1 and Final Time were documented with semi-structured individual interviews. These data allowed us to qualitatively construct three reintegration processes (blocked, downgrading, and sustainable) that describe a large spectrum of workers’ experiences regarding occupational repositioning. Quantitative analyses likewise suggest moderate statistical links between the reintegration process and changes in subjective variables associated with the relationship to work and identity representations. Altogether, the results underline the importance of returning to the labour market in qualified, decent, sustainable work that allows people to have a decent and meaningful personal life. The results also suggest, in keeping with the psychology of sustainability, that interventions should promote occupational and personal enrichment, both at the individual and organizational levels.


Introduction
In the last few decades, the globalization of markets, increasing competition, and the advent of a knowledge-based economy have brought about considerable changes in the production system and organization of work [1]. Human resource practices have become more flexible [2,3]. The types of employment labelled non-standard or precarious [4,5], which are characterized by more unstable ties to work, are becoming increasingly common. A greater number of workers are now on precarious career paths that create numerous impediments to long-term integration into the labour market [6][7][8][9]. Older workers no longer benefit from the protection that accumulated experience in the labour market once provided and are now at greater risk of unemployment [10,11]. In Western countries, the loss of a job at this stage in life can be synonymous with long-term economic precariousness that are "expressive" [49,51,53], "experiential" [47], and "intrinsic" [48,54] are related to people's personal and social development (e.g., self-fulfilment). On the other hand, "instrumental", "economic", and "extrinsic" purposes [47,49,53] concern the more utilitarian aspects of work (e.g., something to do).
The relative centrality of work consists in the value or importance that this sphere of activity is accorded compared to other spheres in life. Work can thus have a primary, equal, or secondary importance in people's lives. Moreover, the relative importance of work can vary according to people's recent work experiences, the occupational context in which they work, and their current priorities or stage in life [8,46,[55][56][57][58]. Finally, it would seem that older workers generally accord higher relative importance to work than do younger workers, and that they also tend to be more committed to the sphere of work than their younger colleagues [47,53,59].
Moreover, confronted with reintegration difficulties, older workers may reduce the relative importance they attribute to work [60,61] and assign a more utilitarian purpose to it. For some, it is a question of serenely accepting the fact that they are getting older, nurtured by the feeling of entering a new stage in life and the desire to invest more energy in activities other than work [62]. For others, the psychological withdrawal can prove to be "painful" when work has always been the main source of personal investment and is still perceived as a wonderful opportunity for personal achievement [13]. It can also be psychologically costly when people must stay in the labour market for several years [63] and feel like they are caught in a trap: too old to be attractive to employers, too young for retirement [34]. In both cases, the decrease in the relative importance of work can be interpreted as a strategy that is set in motion by unemployed older workers to preserve their self-esteem and identity cohesion [60,[64][65][66].

Relationship to Work and Identity Construction
Numerous studies have shown that the more importance people attach to work, the more (1) this sphere of activity represents a source of fulfilment that contributes to their identity and personal development, and the more (2) these people are likely to identify with their work [53,55,[67][68][69]. Accordingly, the unemployed older workers for whom identity destabilization is likely to be the most severe are those for whom work is a central pillar of identity, the place in their life in which they most invest [65,70,71]. Consequently, the meaning that the loss of a job represents can vary from one case to another and affect different aspects of a worker's identity. This identity or "awareness of oneself as a worker" [72] (p. 693) integrates two aspects, namely the occupational identity and social identity, both interacting strongly with each other [73,74].
Occupational identity, that is the "personal definition of oneself for oneself" [73] (p. 389), constitutes the coherent representation that people have of themselves and their uniqueness as workers, with all their skills and values. This representation is expressed in other ways by one's occupational self-esteem [75] and by the feeling of having developed the appropriate skills for an occupation, a profession, or a field of activity [76,77]. As for social identity, that is the "definition of oneself for and by others" [73] (p. 389), it constitutes the workers' identification with a category, an occupational group, or an organization, as well as the recognition they receive from society and their employers [78][79][80]. When the gap is too big between how a worker defines him/herself and how others define this worker, the worker is likely to be confronted with an "identity crisis" [79] and a deterioration in his/her relationship to work.

Identity Loss in Unemployed Older Workers
The consequences of losing one's job in the last third of one's working life can be particularly deleterious at the psychological level and affect one's occupational and social identity [72,81,82]. As regards occupational identity, the self-esteem of unemployed older workers is often affected, their dignity attacked, and their abilities dismissed [83,84], especially when they lose their financial independence due to a lack of employment revenue. Certain authors likewise evoke identity "fragmentation" [85,86] or an identity "rupture" [71] provoked by the gap between the image of experienced, competent people that older workers often have of themselves and that of hard-to-employ people that employers send back to them [87,88]. Moreover, when they lose their jobs, some older workers can feel betrayed by society and undermined in their value system [26,84,86,89,90]. This is particularly the case with those who began working in an economic growth context and who expect to be well treated due to their seniority.
With regard to their social identity, when older workers lose their job, they feel deprived of their worker status that helped to define them and allow them to be recognized by society [91]. Many of these workers have the impression they are no longer useful to a community that does not require their skills [92]. Moreover, the erosion of their network of former work colleagues and the financial precariousness associated with unemployment are likely to marginalize those who are already most isolated [66,70]. For a smaller number of workers, deprived of the socio-temporal framework that work provides and that structures their life in society, losing one's job can result in bona fide social alienation [11].

Labour Market Reintegration as a Psychosocial Transition
At the theoretical level, labour market reintegration is seen as a psychosocial transition, a pivotal moment, a rupture [93] between two stages in life that is likely to provoke major, long-lasting changes in one's representation of oneself and the world [94]. In this sense, this approach emphasizes how people see themselves and integrate these changes into their lives [95]. This approach also implies that transitions take place in the interactions between people and their environments [96], and that these transitions have intra-psychological, interpersonal, and contextual components.
In this approach, transitions are analyzed as a process with three complementary levels [97]. The first level is that of identity transformation and the changes in individual values that can accompany it [98][99][100]. In the present study, for example, this means understanding the extent to which people are led to redefine themselves as workers, depending on whether or not they feel that their skills and experience are acknowledged and that they still have a place in the labour market. This also involves understanding the extent to which reintegration is accompanied by a redefinition of one's priorities in life. The second level involves the actions initiated by workers to deal with changes [101][102][103]. This level looks at the strategies people use to reposition themselves in the job market (e.g., career shift, reconsideration of one's job expectations). It takes into account both their subjective experience and the contexts in which these strategies are implemented. The third level is that of people's (re)construction of meaning, which helps them maintain a stable self in the interactions between their past experiences, new situations, and personal projects [104][105][106]. The third level assumes an understanding of the meaning people give to their new position in the labour market, whether it be separated from or continuous with their previous path. It likewise supposes an understanding of their expectations for the future based on their representations of such things as their age, their stage in their working life, and the meaning of work.
Based on data from an 18-month longitudinal study, this article complements existing studies by examining, from a diachronic perspective and using a mixed-method approach, the labour market reintegration processes of workers who lost their employment at age 45 or over. More specifically, based on a qualitative method focused on people's experience, the article aims primarily to reconstruct these processes according to three dimensions: labour-market reintegration paths, relationship to work, and identity representations. At the same time, using a quantitative method, the article aims to shed light on the statistical links between (1) the reintegration process and (2) the subjective variables associated with the relationship to work and the participants' identity representations.

Participants
Sixty-six adults aged 45 and over participated in the whole longitudinal study. They were recruited through ads in the local press and employment assistance centres situated in Québec City and its surrounding region, in the Province of Québec, Canada.
At the beginning of the study (T1), the researchers met with 110 workers who had been unemployed for less than 12 months. Eighteen months later (TF), the sample comprised 70 participants. The retention rate was thus 64%, which is considered to be acceptable for populations that are difficult to contact [107]. Of the 70 participants who were followed all throughout the study, 61 were selected for the present article; those who at FT were either definitively out of the labour market due to retirement or temporarily due to training were not included.

Material
The data collection was based on a "short-medium duration" longitudinal design [108], that is, 18 months split into four measurement times. For the purposes of the present article, only the Time 1 (T1) and final time (FT), documented with a semi-structured individual interview, were selected. The interviews were conducted in person at T1 (about 2 h in length) and on the phone at FT (about 1 h in length). To make comparisons easier, the two interview grids had the same, four-part structure: (1) an overview of the key events in the career path since the person entered the labour market and an examination of the job loss experience (T1), and an overview of the key events in the reintegration since the job loss (FT); (2) an examination of the employment situation at the time of the interview and the way the reintegration events were experienced and the meaning they took on; (3) predictions about their future possibilities in the job market; and (4) socio-occupational representations (e.g., stages in working life, relative importance of work). All the interviews were recorded and completely transcribed in an anonymized computer file.

Analysis Procedure
The data processing was based on the mixed method embedded design [109], which includes qualitative and quantitative analysis procedures. The aim is to quantify complex and nuanced qualitative data so as to add numerical precision [110]. Congruent with the first objective, the qualitative aspect distinguishes the different processes of labour market reintegration, combining the key events in the processes, some of the workers' representations, and, if need be, the change or stability between T1 and FT of a few of these representations. In keeping with the second objective, the quantitative part made it possible to determine the relations between (1) the identified processes and (2) the change or stability-from T1 to FT-of the variables associated with the relationship to work and the participants' identity representations.

Qualitative Analysis Procedure
The qualitative analysis procedure, which employed thematic content analysis [111], was conducted using an interpretative perspective that focused on the meaning that people gave to their experiences [112]. The analysis, inspired by L'Écuyer's method [113], was conducted in successive steps with an iterative approach [114], as recommended by the APA [115]. Three researchers worked together on the interview analysis using triangulation principles [116] and the "ethics of transparency" [115] (p. 29).
The whole approach can be summarized into two main stages. An analysis grid was constructed in the first stage using an inductive approach based on emerging or reoccurring themes [111], which came up in the participants' comments about their integration into the labour market. A reading of the material collected from 21 of the participants allowed the three analysts to first bring together the identified themes under the three following structural dimensions: (I) the path that reintegration into the labour market took, (II) the changes in their relationship to work from the beginning to the end of this path (T1-FT), and (III) the changes in their identity representations (T1-FT). Due to the iterative data analysis and the addition of interview material, the researchers were then able to divide the three structural dimensions into nine sub-dimensions (A to I), which, put together, made it possible to describe the reintegration processes.
Dimension I: (A) sequence of events and changes in the reintegration path, (B) occupational situation after 18 months (FT), (C) main obstacles observed in the reintegration path, (D) occupational self-esteem at the end of the path (FT), and (E) expectations for their future in the labour market (FT).
Dimension II: (F) changes or stability in the relative importance of work with respect to other spheres in life (primary, equal, or secondary importance) and (G) changes or stability in the dominant purpose of work (expressive or instrumental).
Dimension III: (H) changes or stability in the representation of age in the labour market (seen as a strength, limitation, or obstacle for remaining in the labour market) and (I) changes or stability in representation of the stage in one's working life (feeling of being at the midpoint or the end of one's working life).
The three analysts then proceeded to separate the nine sub-dimensions into finer categories. By adding data from the interviews to the other 40 participants, the definitions of the nine sub-dimensions and their categories were reworked and fine-tuned using the "self-correcting process" [115] (p. 27). Two analysts divided the coding of the extracts; when difficult cases arose, they were arbitrated by a third analyst. In this way, all the data relative to the reintegration processes of the 61 participants were categorized using the analysis grid shown in Table 1. This grid allowed us to draw up a general portrait of labour market reintegration as experienced by the participants.    The second main analysis stage consisted in qualitatively reconstructing, in the form of a narrative discussion [116] (pp. 254-255), the different processes of labour market reintegration. To achieve this, the three researchers first began a content analysis of the collected material so as to understand the interconnections between the dimensions, sub-dimensions, and categories of the analysis grid for each participant. An inter-case analysis, that is between the participants, was then conducted [114], which allowed us to bring together processes presenting strong similarities. By grouping together similar cases and by analyzing the interconnections between the dimensions, sub-dimensions, and categories of the analysis grid, a typology composed of three reintegration processes was then constructed. Using a progressive and iterative approach, this typology was gradually improved through exchanges between the three researchers with the analysis triangulation method [116]. Finally, by interpreting and integrating the relations between the sub-dimensions and categories of the analysis grid, each of the three processes was described in detail. Two of the analysts divided up the 61 study cases in order to place them in one of the three identified types of processes. The most difficult cases were arbitrated by a third analyst in the form of blind tests.

Quantitative Analysis Procedure
Complementary quantitative data processing was conducted to enhance our understanding of the reintegration process. Using SPSS 24 software, a non-parametric procedure that is an inter-group comparison (i.e., between the three reintegration processes) was conducted on four sub-dimensions of the qualitative analysis grid: F, G, H, and I. So as to avoid too large a dispersion of the participants, these four sub-dimensions were processed dichotomously.
Furthermore, considering the limited number of participants and several convergences between the first two types of processes in the typology developed here, these two processes were combined into a sole category (increasing fragility). We then calculated the frequency distribution resulting from the comparative analysis of the two reintegration processes thus identified and each of the four dichotomous sub-dimensions. Pearson chi-squared tests were used to verify the existence of relations between the processes and sub-dimensions, and odds ratio calculations were used to estimate the size of the possible relations [117].

Results
We first present the three qualitatively reconstructed integration processes. We then present the results of the complementary statistical analyses.

A Blocked Reintegration Process Marked by of Sense of Deadlock
There were 12 participants in the first process, that is, 20% of the sample, split equally between men and women. Three quarters of these participants were in the two highest age brackets (51-55 years old, n = 4; 56 years and over, n = 5). Half had university degrees, four others had community college diplomas, and the other two had high school diplomas.
All of these participants experienced a "chaotic" reintegration marked by a sporadic presence in the labour market. Three of these participants found no work during the whole length of the study. Most (n = 9), however, found at least one job during the process. These jobs were far from meeting their expectations and required compromises that were not very useful, since these participants all found themselves once again unemployed. Some had to accept a string of precarious jobs (3 to 5 in 18 months) that paid poorly and had no connection with their skills. This was the case for a former community organization director who had worked in, in the same month, two part-time jobs for which he was over-qualified as an administrative support agent, before finding himself once again unemployed. Others succeeded in finding a fairly steady job that was somewhat associated with their skills, but had to take a sizable drop in advantages (e.g., salary, appeal of the tasks). After a year, however, they lost these jobs that they thought were stable, "a hard blow for my pride, especially at 55". One university graduate in her fifties recounted the following: "It took me 18 months to find another job. This job wasn't at all what I had done before, not at the hierarchical level, at the responsibilities level, or the salary level. ( . . . ) I also had to make sacrifices: ( . . . ) I spent the year driving [between my house and my workplace] ( . . . ). I was certain they were going to renew my contract for another two years, but they let me go!" Eighteen months after T1, all of these workers were still looking for a job, having to endure both occupational downgrading and precariousness. They thus had to deal with severe financial precariousness, especially when they did not have the right to employment insurance. The contrast with their former work situation was even more striking, given that most (n = 10) had had a stable, permanent job and half had held management or administrative positions.
Having to constantly deal with "doors that close in your face" and "never-returned phone calls", all of the participants reported having encountered sizable obstacles. Among them, a little more than half (n = 7) said they were confronted with obstacles that were both individual (e.g., overly specialized expertise) and contextual (e.g., hiring discrimination because they were perceived to be too costly for employers).
Feeling disheartened because neither their skills nor their experience were acknowledged, almost all these participants (n = 11) had a negative occupational self-esteem at FT. They wondered about their place as workers, as stated by this participant: "You lose your confidence about being able to find a new job but also about your personal worth. ( . . . ) It's as if you're worth less because you can't find a job and because you're unemployed even though you're still young enough to work!" At FT, all the participants expressed uncertainty or resignation about their future in the labour market. Considering that the situation "could not be worse" than the one they were in, they said that they had no other goals than to remain in the job market and meet their needs in the short-, mid-, and even long-term. A university graduate in her early fifties told us that she was ready to accept "just about anything": "I hit two walls one after the other. It's like having a serious illness; you don't see things the same way anymore. I keep telling myself, "As long as I make enough to get by, it's okay!" Repeated failure to reintegrate the labour market and the sense of an impasse that accompanied this failure led many of these workers to reconsider their relationship to work. "It's like something has been torn apart," emphasized one female participant. The change was especially noticeable as regards the relative place of work in their lives. In fact, whereas at T1, more than half of them (7/12) saw work as a priority in life, at FT, these people, with only one exception, relegated it to a place that was equal or secondary to the other spheres of activities. This distance they had adopted regarding work was seen as a stop-gap measure, an enforced "letting go," as one woman reported: As regards the main purpose of work, most of the participants in this process (9/12) maintained the same position during the 18 months of the study (expressive, n = 4; instrumental, n = 5). The three others went from a view of work as a privileged place where one "fulfils oneself" to a "pay-the-rent" view, in the words of one female administrator. Consequently, the instrumental purpose was in the majority at FT (8/12), whereas the expressive purpose dominated at the beginning of the study (7/12).
As concerns the representation of age, half of the participants maintained a negative view from the beginning till the end of the reintegration process. For a third of the participants, however (4/12), their representation of age evolved negatively: age became an obstacle to occupational progress at FT, whereas it was seen as a strength at T1. One female participant deplored the fact that employers took advantage of her years of experience without paying her accordingly and treated workers her age like "cheap labour".
As for their representation of what stage they were at in their working life, almost half of the participants (n = 5) went from the impression that they were at the midpoint, where it was still possible to develop work projects, to the belief that it was basically over, that the main part of their working life was behind them. As one administrative executive summarized, "All the things I like to do are over. It's too late and I can't do anything about it".

A Downgrading Reintegration Process Marked by the Sense of Being Undervalued on the Labour Market
In this second process, which comprised 20 participants (32%), women represented close to two-thirds (n = 13) of the group. Unlike process 1, most of the workers (80%) were concentrated in the two youngest age brackets of the sample: 45-50 years old (n = 9) and 51-55 (n = 7). Their educational level was heterogeneous: the number of participants with a high school diploma (n = 8) was similar to those with a university degree (n = 9), whereas half this number (n = 4) had a community college diploma.
These participants had for the most part held one or two jobs during T1-FT. Of these, half (n = 10) were employed for most of the period under study ("winding" path); a slightly smaller number (n = 7) experienced more employment/unemployment intermittence ("chaotic" path).
All the workers in this process had a job at FT, but their situations had substantially degraded since T1. Close to two-thirds (n = 13) had experienced occupational precariousness, going from a stable job (e.g., permanent, renewable contract) to casual or on-call jobs. Slightly more than half (n = 11) were obliged to accept part-time jobs, which led several to hold jobs in heterogeneous activity sectors (e.g., secretarial and restaurant). Moreover, the job held by almost one out of two participants (n = 9) at FT was in a quite different activity sector from that held at T1. What is more, a third (n = 7) underwent occupational downgrading, quite often going from a technician-level job to a labourer-level job. Finally, close to two thirds (n = 13) experienced considerable financial precariousness. (By financial precariousness, we mean a loss of at least C$250 per week (gross income). This represents an annual decrease of some $13,000 in a context in which the median employment income for all the Province of Québec was $35,746 [118].) The fact of finding a precarious job for which they were over-qualified was seen by these workers as a consequence of the obstacles they encountered. For instance, they noted the importance of individual obstacles (n = 7) related to their training (e.g., lack of certification) or to health problems. Others (n = 7) mentioned that the individual obstacles were added to contextual obstacles, especially ageism. One woman summarized it thus, "Yes, they acknowledge my skills but they always say that I'm missing something for the job".
The majority of the participants in this process (n = 15) expressed, like those in process 1, a negative occupational self-esteem at FT. Even though they were employed, they felt underappreciated as workers and unacknowledged by employers. They felt that they "could have been put to better use" in the words of one female participant. In this situation, most of them were looking, more or less actively, for another job with more "decent" conditions, given the stage they were at in their working life (e.g., regular, daytime work hours) and their skills. This was the case for a manager in his fifties whose new job at FT was fairly stable (renewable contract) but whose tasks were those of a subordinate.
As for their future in the labour market, most of them expressed serious doubts about any real possibilities of improving their situation.
As concerns the relative importance of work, it decreased for half of these participants (n = 10) over the 18 months. Most often at T1, work was on an equal footing with the other spheres of life but in a secondary position at FT. It also went from a primary importance to an equal or secondary importance with respect to the other spheres in life. Moreover, close to half the workers belonging to this process (n = 9) readjusted what they considered to be the main purpose of work, this occurring on a much larger scale than in the other processes. Seen as an end unto itself and a source of "accomplishment" (expressive purpose) at T1, work by FT became a way of ensuring one's subsistence and financial independence (instrumental purpose). As expressed by a former manager who reconverted, "for lack of anything better to do," into a carpenter, work lost its meaning and was nothing more than a "necessary evil [because] you have to work in life".
What is more, with little hope of finding a job in keeping with their occupational skills, almost half of the participants (9/20) stated there was a negative evolution in their representation of age during the 18 month period. Some who, at T1, saw it as a factor in renouncing new work projects (limitations), saw it at FT as an outright obstacle to their occupational progress. The others, who initially saw their age as a strong point in their labour market reintegration, at the end saw it as a limitation or even an obstacle.
Likewise, slightly more than a third of the participants (n = 7) felt they were at the end of their working lives at FT, whereas they were confident they were at the midpoint of their working lives at T1. This was expressed by a former sales representative in his mid 50s, humiliated and obliged to make a living as a janitor: "I think my working life is behind me now. ( . . . ) Before, I had a much better position: I was a lot more than a janitor! I would have liked to finish out my working life doing something other than cleaning buildings."

A Sustainable Reintegration Process Marked by a Sense of Enhanced Integration in the Labour Market
This third process grouped together 29 participants, that is, close to half (48%) the sample. As in the second process, women were in the majority (n = 20; 69%). More than 85% were concentrated in the youngest age brackets: close to one out of two (n = 14) were from 51 to 55 years old, and a little more than a third (n = 11) were from 45 to 50. Most had a high school (n = 12; 41%) or community college (n = 10; 35%) diploma.
The reintegration paths these people experienced were either "linear" (n = 20) or, more rarely, "winding" (n = 9), with little or no intermittence between jobs. In the case of the linear paths, most of the participants joined organizations early on in which they were still working at FT. The winding path involved people who chose occupational retraining at T1.
At the end of the 18-month-long study, the majority of the participants in the third process had found a job in-keeping with their aspirations and of the same qualification level as the job that they had lost. Almost all (n = 27) found themselves in stable jobs of the same status as at T1 (i.e., permanent or contractual). Most of the participants (n = 23) did not experience financial precariousness and earned a salary at FT comparable to that of T1.
While the return to a stable job was not without its difficulties, most of these workers (n = 22) considered that they had not encountered major difficulties since, in the end, they had "found their place".
In this context, all of them considered their occupational self-esteem to be positive at FT: their self-confidence as workers was reinforced by the fact that they had found jobs that provided them with acceptable working conditions and long-term integration into the job market.
What is more, three quarters of these workers (n = 22) saw their future working life optimistically and said they were confident that they would keep their job for several more years. This job was often seen as a "good job to finish your career with" in the words of one educational consultant in her early fifties. As for a few other participants (n = 7), they accepted, fairly serenely, to finish their working lives with the job they had found, even though it did not exactly correspond to what they would have preferred. This was the case for a woman in her mid-forties, with a community college teaching diploma, who found a job in her field, but only for 28 h a week: "Of course I would have liked to work closer to home and to have a few more hours per week ( . . . ). But the way it is, if nothing changes [and there are no more job offers in the next few years], I can see myself where I am now, and that's okay." Unlike the other two processes, most of these participants (n = 22) attributed the same relative importance to work all throughout their reintegration path, which in general was equal to the importance of other spheres in their life. As one woman explained, "Work has never been and never will be the priority in my life, but it's still up there with my family and health". Moreover, in the few cases where the relative importance of work decreased (n = 6), this activity remained a sphere of significant investment at FT (of a primary to equal importance).
As concerns the dominant purpose attributed to work, slightly more than one participant out of two (n = 16) maintained their perception of work as a source of "accomplishment", a "pleasure", or a "service" rendered to society (expressive purpose) all throughout the T1-FT period. The attribution to work of more instrumental values was a rare phenomenon (4/29).
Regarding their representation of age, close to one participant in three (n = 10) continued all throughout their reintegration path to associate age with occupational "challenges" that were still within reach (strength). One man described it as follows: "Being over 50 is not a reason to say, 'You can't find another job.' On the contrary, there are always jobs available whatever your age. You've got experience. You'll always manage to find something." Moreover, a comparable proportion (n = 9) developed a more positive view of their age (from a limitation to a strength) between T1 and FT, which was reinforced by the feeling of having successfully returned to the labour market, sometimes above and beyond their hopes.
As for their representation of the stage in their working life, it remained the same from T1 to FT for more than half of the participants (17/29). Either they already saw themselves as being at the end of their working life at T1 (n = 11), or, to a lesser extent, they continued to see themselves as being at the midpoint. The others (n = 12) went from a feeling of being at the midpoint (T1) to that of being at the end of their career (FT). The latter reported being satisfied with their repositioning in the labour market and with their accomplishments in their working life, as well as being happy to undertake the last leg of their working life.

Complementary Statistical Analyses
Regarding objective 2, a moderate statistically significant relation was observed between the types of reintegration process and changes in three of the four sub-dimensions from T1 to FT (Table 2): the relative importance of work with respect to the other spheres of life (χ 2 (1) = 6.814, p < 0.01), the dominant purpose of work (χ 2 (1) = 4.419, p < 0.05), and the representation of one's age in the labour market (χ 2 (1) = 5.696, p < 0.05). However, these latter results should be interpreted cautiously because of their moderate level of significance. Table 2. Change or stability in the relative importance of (1) work with respect to the other spheres in life, (2) the dominant purpose of work, (3) the representation of age in the labour market, and (4) the representation of the stage in one's working life according to the type of reintegration process. n.s.: not statistically significant; *: p < 0.05; **: p < 0.01. (a) including the rare participants (n = 4) whose dominant purpose went from instrumental to expressive between T1 and FT; (b) four participants were excluded because their responses could not be coded; (c) including the rare participants (n = 3) whose feeling went from being at the end to the mid-point of their career between T1 and FT.

Type of Reintegration Process Evolving Variables
The odds ratio method [117] indicates that, among the workers who experienced increasing labour market fragility as compared to those whose reintegration was sustainable, the probability over the 18-month period of stepping back from work with respect to the other spheres of life was 4.3 times greater; that of instrumentalizing work, that is, making it a means to an end, 3.8 times greater; and that of developing a negative perception of one's age in the labour market, 5 times greater.
On the other hand, the change in one's feeling that one is at the midpoint of one's working life to that of being at the end of it did not appear to be significantly related to the type of reintegration process (χ 2 (1) = 0.001, n.s.). It is worth noting that none of the four variables examined appeared to be significantly related to the participants' chronological age.

Discussion
The results of the study presented here shed light, on the basis of statements from interviews with unemployed workers 45 years old and over, on three labour market reintegration processes. They also suggest some moderate statistical relationships between these processes and three variables associated with the relationship to work and the workers' identities.

Occupational Repositioning in the Labour Market: Contrasting Experiences
Echoing previous research that emphasizes the difficulties that older unemployed workers have in Canada and in Québec in their attempts to re-enter the job market (e.g., [24,25,30]), our results showed that, after 18 months of study, less than half of the workers that we met had found a stable job in keeping with their accumulated experience and qualifications (process 3). Conversely, the other half of the participants were still without work (process 1) or confronted with financial and occupational precariousness, downgrading, and underemployment (processus 2), a reality which numerous authors have observed (e.g., [26,27,29,66,71,119]). Moreover, examining, from a diachronic perspective, the reintegration of unemployed older workers as a psychosocial transitional process [94,95,97], our study adds to existing knowledge by relating the sequence of major events in the three reintegration processes identified here with the way the workers experienced and gave meaning to them. The three types of processes accordingly describe contrasting experiences that can be located on a continuum.
At the one end of this continuum, process 1 brought together people who were unemployed following a sequence of entries into and exits out of the labour market, without any real control of the key events on their path. The reintegration path was thus seen as a genuine "failure" which, in addition to the financial losses, was seen as an "irreversible" rupture that was above all unexpected in their career path and that led them to doubt themselves and their ability to turn the situation around. At the other end, the workers in process 3 repositioned themselves in jobs that were comparable to those they had lost, and this after a fairly short time without any significant obstacles. The experience was interpreted here as an "incident" in their career path which, far from being perceived negatively, contributed rather to strengthening their self-esteem as workers and their confidence in their future career path. Between these two ends of the spectrum, the participants comprising process 2 had jobs, but their reintegration path was marked by numerous obstacles and periods of unemployment. The reintegration process was perceived as a genuine "downgrading" given the considerable financial losses they encountered and their decline in job status. The participants said the experience was like sliding down a slippery slope, a downward career path over which they had little or no control. More fundamentally, it was interpreted as an attack on their dignity and the loss of their place in the labour market.
These results illustrate the notion according to which re-entering the labour market after losing a job at 45 or older constituted, for the interviewed workers, a pivotal moment in the stages in their life [93,94]. They also show the importance of taking into account, so as to grasp its complexity, how workers experienced this reintegration process over time and gave it meaning. The results demonstrate, moreover, that this process created a strong destabilization in the participants' identity and a feeling of discontinuity, and even a rupture in their career path in processes 1 and 2. Conversely, for the participants in process 3, their reintegration would seem to have contributed to a strengthening of their occupational and social identity and their sustainability in the labour market.

The Two Faces of Occupational Repositioning
As concerns occupational repositioning, it is clear that there are some nuances to be made at an objective level about the distinction between a return to work versus recurrent unemployment, particularly with regard to the workers' subjective experience. Indeed, while only 12 of the participants (19%) were unemployed at the end of the study (process 1), a large proportion of those who returned to the labour market (40%, process 2) were faced with precarious employment that was particularly difficult to handle. Hurt not only at the occupational and financial level but also at the psychosocial level, these older workers expressed, like those who were unemployed, a psychological malaise and a feeling of a "loss of self" [11,89], a social uselessness [92]. These results are worthy of researchers' and practitioners' attention given that a greater number of workers now navigate in career paths marked by precarious employment and come with more complex guidance needs [4][5][6][7][8][9]90,105,120]. These results contribute at the same time to enhancing our comprehension of the deleterious consequences, for workers 45 and over, of not being able to reposition themselves in the labour market. That is to say, to meet not only minimal standards in terms of working conditions (e.g., safety, social advantages) but also to satisfy their aspirations for a decent job and a career that promotes effective and sustainable well-being (e.g., [2,3,[37][38][39][41][42][43][44]).

Relationship to Work and the Reintegration Process: A Complex Interaction
Given the links between the reintegration experience and the relationship to work, our results accordingly indicated that a third of the participants reported a decrease in the relative importance they accorded to work. The statistical analyses showed that this decrease involved more destabilized workers (processes 1 and 2), who were experiencing the most difficult reintegration situations. This result enhances our understanding of the variability of the relative value attributed to work based on recent experiences in this activity sphere (e.g., [46,56,57]). Additionally, while the greater distance that participants put between themselves and work can be interpreted as a strategy to preserve their occupational identity, particularly their self-esteem (e.g., [60,61]), our analyses also showed that this "choice" was often reported by those whose self-esteem had declined the most at FT. This result suggests that greater nuance is required to interpret the distance that workers put between themselves and work, at least in terms of its effectiveness in protecting themselves. Indeed, this distance could also illustrate how the decrease in the importance accorded to work contributes to a lowering of one's own self-esteem rather than protecting it, thereby confirming the observations of several authors, particularly in cases where people had previously accorded a great deal of importance to their work [66,68].
With respect to the relations between repositioning oneself in the labour market and changes in the dominant purpose of work, our analyses suggested that more than one third of the participants in processes 1 and 2 went from an expressive purpose to an instrumental one (vs. only 14% in process 3). However, the qualitative analysis reveals that this transition was associated more with individuals in process 2, those whose occupational repositioning led to a downgrading professional path. Thus, a possible hypothesis could be that the instrumentalization of work translates a work's loss of meaning more than a deliberate strategy allowing the study's participants in process 2 [34,63] to deal with the difficulties of reintegration. Although this result has to be interpreted cautiously and needs to be confirmed with a larger sample and with different populations of workers, it deserves the attention of researchers and practitioners given that remaining in a work situation that "makes no sense" can have considerable impact on the psychological health of workers [27,66,82], but also because the participants in process 2 belonged to the youngest age category and will have to stay in the job market for several more years. From a psychology of sustainability perspective, this result stresses the necessity of acting to promote meaningfulness in organizations using a primary prevention intervention approach [3].

Representations of Oneself as a Worker and the Reintegration Process: Mixed Interactions
Our statistical results revealed that the negative evolution of the representation of age was more often a problem for the destabilized workers in processes 1 and 2, as was the relationship to work. This result leads us to suggest the hypothesis of a significant long-term relationship between, on the one hand, downgrading and the loss of social status and, on the other, the destabilization of the occupational and social identity of the participants in our study. The comments of the workers in processes 1 and 2 gave particularly evocative illustrations of the difficulties involved in both returning to the labour market with a job in keeping with one's skills and in composing with the status of an "unemployed older worker", that is, someone who is "difficult to employ". These two elements contributed to both a feeling of loss and identity deconstruction [64,83,84,88]. Conversely, the positive evolution in the representation of age primarily concerned workers whose attachment to their work was reinforced (process 3). Even if this result cannot be generalized and requires to be validated with more data, it seems to support the idea according to which, when people find jobs during their reintegration that have meaning for them and in which they feel acknowledged, their occupational and social identities are not only likely to be preserved but also to be reinforced [72,76,80,121].
As for their representation of the stage they were at in their working life and the change in the representation according to the process, our results did not reveal any statistically significant difference. However, the qualitative analysis of their statements helps us to better understand how the transition from a feeling of being at the midpoint in one's working life (T1) to that of being at the end (FT) can represent two quite distinct realities. On the one hand, people who experienced a stronger attachment to their jobs (process 3) seemed to correspond to the definition that Marbot [62] gave to the feeling of being at the end of one's working life, that is, a serene acceptation of a new stage in one's life at work and a corresponding redefinition of one's priorities in life. Generally proud of what they have accomplished in their working lives, the participants said that they simply wished to "hold onto what they had gained" and chose to adjust their expectations and life goals accordingly.
That being said, our analysis of the destabilized participants' statements (processes 1 and 2) illustrated a whole other reality. Moving from a feeling of being at the midpoint of their working life to that of being at the end appeared to be the consequence of an identity impasse. This impasse was provoked by the gap between the representation that the participants had of themselves and what they had to offer at T1, and that which society was sending back to them over the 18 months. Our data illustrates the tension between a social identity for oneself and for others, as well as the destabilization that this tension can lead to when the gap between these two dimensions of social identity becomes too large [34,79,87]. Moreover, taking into account the medium-term effects of the reintegration on the workers' identity, these results add to the scientific assessment of the way in which unemployment and occupational downgrading can generate a feeling of social uselessness in seniors and even marginalization (e.g., [11,91,92]).

Limitations
Our study has several limitations. First, because our sample includes a relatively small number of workers, the interpretation of our results must be done with a lot of caution. In addition, although the sample is quite diverse (for example, in education), it is not statistically representative of the population of the unemployed aged in Quebec City, the province of Quebec, or Canada. Thus, it is difficult, if not impossible, to generalize our results to all Canadian workers.
Besides, we did not take into account the context of the Canadian labour market. It would probably have been interesting to do so but this would have wildly overpassed the goals of the present study.
Finally, the present study has operationalized the relation to work and the representations of identity using a limited number of variables, whereas these notions are multidimensional and complex as mentioned in the literature (for example, [47,49]).

Implications for Research
The present study could be extended at various levels. First, a larger sample would allow us to verify the validity of the three reintegration processes identified here and to confirm the statistical relations observed between the types of processes and the change in the psychosocial variables studied. It would also enable us to shed light on the role of socio-biographic variables (e.g., age, education) in the statistical relations observed. Specifically, it would be worthwhile to compare workers aged 45-60 versus those aged 60 and over so as to better understand the effect of aging in the labour-market reintegration process and the way it is experienced. Likewise, a more diversified sample in terms of age would have allowed us to determine whether the statistical relationships observed also concerned younger unemployed workers.
Second, it would be appropriate to pay particular attention to the multidimensionality of the notion of relationship to work and of identity representations in future research. Concerning the relationship to work, it would be worthwhile to have regard the absolute value of work [122,123], as well as the representation of a decent job and a sustainable work life from the participants' perspective [3,19,38]. Concerning identity, aside from a more precise operationalization of occupational and social identities, it would be interesting to compare the changes in the representations of oneself as a worker with changes in the self-representations in other spheres of life. Indeed, given the interdependence of life roles, occupational changes are likely to have an impact on work life but also on life outside work, in particular on one's personal and family life [100].
Third, it would be relevant to study the labour market reintegration processes over a longer period than the 18-month period of the study in order to take into account their sustainability from a psychological perspective, as well as to analyze to what extent the scarring psychosocial effects of job loss at this stage of life might last.

Conclusions
Our results suggest some reflections concerning organizational policies for the management of older workers. The latter would benefit from taking into account the particular needs of these workers, especially the expectations they pursue in their working lives according to the stage of their lives (e.g., mid-career or end of career). They could also pay particular attention to the wealth of skills that many can offer to support the sustainable development of the organization. In short, older worker management policies that would support a better linkage between the particular needs of these workers and those of the organization are likely to generate greater individual well-being at work and, in turn, to contribute to the sustainable development of organizations.
Our results also suggest some reflections regarding individual counseling with workers aged 45 and over in their search to reintegrate the labour market. It would be helpful to help people develop their personal and professional enrichment strategies (e.g., [3,37,38,40,41,124]). It would likewise seem important to develop interventions that take into account the priorities and living context of unemployed older workers, what they expect regarding work, and how they conceive their future in the labour market. Moreover, practitioners should be attentive to the nature of the identity wound in question, the meaning accorded to it, and the strategies to be developed to limit its long-term impact. In order to re-contextualize this meaning, they could pay particular attention to the circumstances surrounding the loss of employment and to what that meant for the workers. Finally, it could be highly relevant to use an intervention based on a narrative approach [8,[125][126][127][128], as promoted by the «meaning paradigm» [3]. With such an approach, the accompanied people are invited to think about the meaning of their past experiences in order to choose what they want for the future. Through their stories, it is also possible to identify key elements at both the individual and organizational levels that may promote sustainable development and well-being.
Author Contributions: G.F. and H.Z. wrote the paper jointly. G.F. is the scientific director of the research project on which the present article is based. She is the recipient of the research funding allowed to the project, and she managed all its steps from its design to the investigation and data analysis. J.M., a specialist of psychosocial transitions, contributed to the writing of the article by commenting, in detail, on a draft version of the whole article. C.G. participated in the empirical investigation and conducted the analysis of the data collected, along with G.F. and H.Z.