When is the grass greener on the other side? A longitudinal study of the joint effect of occupational mobility and personality on the honeymoon-hangover experience during job change

Previous research shows that job satisfaction often increases sharply upon initial entry into the new job and gradually falls back to the baseline level over time. In this study, we propose that this ‘ honeymoon-hangover ’ pattern is affected by both the direction of occupational mobility and the individual's personality in terms of extraversion and neuroticism. Drawing on the British Household Panel Survey that followed 10,000 individuals annually for 18 years, this study shows that only those who move up the occupational class ladder experience significant ‘ honeymoon ’ effects, while those who move downwards experience dissatisfaction that lasts for several years after the transition. While the positive effect of upward mobility is not amplified by extraversion, the negative effect of downward mobility is exacerbated by neuroticism. This study highlights the importance of taking into account both situational and dispositional factors for understanding the long-term impact of career change on subjective well-being.


| INTRODUCTION
Research on job change has revealed a 'honeymoon-hangover' pattern during the turnover process, which refers to the fact that job satisfaction typically peaks initially following a job change but subsequently falls back to the baseline level over time (Boswell, Boudreau, & Tichy, 2005;Chadi & Hetschko, 2018). We argue that this pattern may be more nuanced than previously revealed because the effect of job change can vary across situations and individuals. People's reactions to turnover are likely to be influenced by both the nature of job change and their personality traits because the former determines the objective changes to their work environment and the latter affect the subjective appraisal of the transition. The aim of this study is to advance our understanding of the long-term impact of job change on subjective well-being by considering how situational and dispositional factors jointly shape post-turnover job satisfaction trajectory.
To capture the nature of job change, we focus on the direction of occupational mobility, a type of job change that involves moving across occupational boundaries-'a change to a work position in a different general field of work in which the major tasks, activities and responsibilities are different in nature' (Breeden, 1993, p. 33). As occupational mobility involves significant modifications to job content, skills, and work routines, it has a greater impact on job satisfaction compared with within-occupational job changes where individuals move to different organizations but continue to perform similar job tasks (Zhou, Zou, Williams, & Tabvuma, 2017). Occupational mobility therefore serves a desirable indicator to unpack the impact of job change on post-turnover well-being. Since people generally have selfimprovement desires (Sedikides & Strube, 1997), upward occupational mobility is often seen as a success at work that can lead to a boost of subjective well-being at the time of turnover, whereas downward mobility tends to result in feelings of deprivation and disappointment.
However, these initial emotional responses to job change are likely to wear off over time as set point theory suggests that fluctuations in subjective well-being around life events are transient and individuals will return to their baseline well-being after they have adapted to the changed life circumstances (Headey & Wearing, 1989;Larsen, 2000).
Our second main argument is that people react differently to occupational mobility due to their personality traits. Drawing on Gray's (1981Gray's ( , 1987 biopsychological theory of personality that emphasizes individual differences in sensitivities to reward and punishment stimuli, we propose that individuals high in extraversion (those who tend to focus more on gains and pleasurable experiences) will demonstrate greater reactions to upward occupational mobility while those high in neuroticism (those who tend to focus more on losses and threatening experiences) will demonstrate greater reactions to downward occupational mobility. As it takes longer time for those who react more strongly to an external event to adapt to the change (Fredrick & Loewenstein, 1999;Lucas, Clark, Georgellis, & Diener, 2003), we expect that individuals high in extraversion or neuroticism will take longer time to return to their baseline well-being following occupational mobility compared with their introverted or emotionally stable counterparts.
This study contributes to the literature in three ways. First, by examining the direction of occupational mobility, it reveals variations in job satisfaction trajectory that deviate from the 'honeymoon-hangover' pattern revealed by previous research, highlighting the importance of job content in shaping post-turnover well-being. Second, by taking personality into account, this study shows that individuals can demonstrate idiosyncratic job satisfaction trajectories during the turnover process, a phenomenon that has rarely been examined in previous research.
Finally, our study enriches set point theory. Diener, Lucas, and Scollon (2006) have called for a revision of set point theory that incorporates individual differences in the rate and extent of adaptation to various life events. Our study answers this call by revealing substantial individual differences in adaptation to career change.

| Direction of occupational mobility and job satisfaction trajectory
Occupation, defined as 'a category of functionally similar jobs' (Grusky, 2005, p77), represents an important indicator of one's socio-economic position. A substantial body of research shows that highly skilled occupations provide both higher levels of economic rewards (Kalleberg, 2011;Williams, 2013;Zou, 2015) and intrinsic rewards in terms of task variety, job autonomy, skill development opportunities and participation in organizational decisions (Boxall & Macky, 2014;Charlwood, 2015;Felstead, Gallie, Green, & Henseke, 2019;Holman & Rafferty, 2018;Kalleberg, 2011;Williams, Zhou, & Zou, 2020). As such, upward occupational mobility can be an important means for individuals to improve their well-being. Physical well-being is maximized through the satisfaction of material needs, and social well-being is maximized through the gratification of the desire to gain approval, respect and admiration from others. Moving up the occupational ladder can improve both dimensions of well-being by increasing job resources and eliciting positive perception from others (Hadjar & Samuel, 2015).
In addition, upward occupational mobility can enhance skill utilization and strengthen one's sense of competence, which satisfies the inherent human desire to feel effective in interacting with their environment (Deci & Ryan, 2000).
By contrast, downward mobility can adversely affect an individual's subjective well-being. Besides its negative impact on access to economic and social resources, downward occupational mobility can also lead to feelings of deprivation stemming from the lack of opportunity to fully utilize one's skills (Crosby, 1976), a major source of dissatisfaction at work (Morrison, Cordery, Girardi, & Payne, 2005).
Downward mobility often results in overqualification, a condition in which an individual has more skills and experience than required by the job (Erdogan, Bauer, Peiro, & Truxillo, 2011;Maynard, Joseph, & Maynard, 2006). Failure to obtain a job at one's expected level can lead to feelings of disappointment and frustration (Johnson & Johnson, 1999;Wu, Luksyte, & Parker, 2015). Compared with upward and downward occupational mobility, lateral mobility in which people change their jobs within the same occupational class is likely to have least impact on subjective well-being because it incurs relatively minor alternations to one's current socio-economic status.
To gauge the impact of occupational mobility, we focus on job satisfaction because it reflects a positive emotional state resulting from the experience of work (e.g., Locke, 1976) and is more relevant to vocational behaviour than global measures of subjective well-being such as life satisfaction and mental health. For example, while low levels of job satisfaction predict absenteeism and turnover (Clark, Georgellis, & Sanfey, 2012;Freeman, 1978;Green, 2010), positive trajectories of job satisfaction often reflect career success (Judge & Hurst, 2008;Wu & Griffin, 2012). In addition, job satisfaction is widely regarded as an important measure of employee well-being (e.g., Warr, 1999) and is the key variable used to capture the 'honeymoonhangover' effect during job change (e.g., Boswell et al., 2005;. Based on the discussion, we propose the following: Hypothesis 1. Upward occupational mobility leads to a significant increase in job satisfaction upon turnover.
Hypothesis 2. Downward occupational mobility leads to a significant decrease in job satisfaction upon turnover.
While occupational mobility can induce job satisfaction fluctuations at the time of turnover, its impact is likely to dissipate over time.
Set point theory argues that subjective well-being is largely determined by genes and remains generally stable over the life course (Headey & Wearing, 1989;Larsen, 2000;Lykken & Tellegen, 1996).
Although external events may shift subjective well-being in the short term, individuals will gradually regress to their baseline after they have adapted to changed life circumstances, a process described as the 'hedonic treadmill' (Brickman & Campbell, 1971). Empirical research has provided support for set point theory with respect to a wide range of life events such as marriage, birth of child, divorce, bereavement, lottery winning and debilitating injuries (Brickman, Coates, & Janoff-Bulman, 1978;Clark, Frijters, & Shields, 2008;Lucas et al., 2003).
Although questions about the universality of this process remain, 1 by and large existing evidence shows that the adaptation phenomenon is widespread and human beings often demonstrate considerable resilience in the face of adversity and misfortune (Bonanno, 2004).
Following set point theory, we propose that although upward occupational mobility initially generates a significant 'honeymoon' effect due to the upgrading of one's socio-economic status, this effect will gradually wear off after individuals have adapted to their new job tasks, responsibilities and resources. In other words, the initial novelty will turn into regular routines after people have become more familiar with their work environment. Similarly, downward occupational mobility will initially generate significant dissatisfaction due to the loss of job resources and social status, but with sufficient time, people will return to their baseline well-being after they have come to terms with the new reality. For example, individuals may gradually accept the situation by justifying the value of performing the new job (Lepisto & Pratt, 2017) or crafting their work according to their skills and preferences (Berg, Wrzesniewski, & Dutton, 2010). We thus propose that Hypothesis 3. The initial increase in job satisfaction upon upward occupational mobility will dissipate over time.
Hypothesis 4. The initial decrease in job satisfaction upon downward occupational mobility will dissipate over time.

| Individual differences in reactions to occupational mobility
Besides the direction of travel on the occupational class ladder, the impact of job change can be also affected by an individual's personality. In this study, we focus on extraversion and neuroticism. As job satisfaction reflects a work-related positive emotional state, individuals high in extraversion, who tend to experience more positive emotions (McCrae & Costa, 1987), are likely to experience higher job satisfaction, whereas individuals high in neuroticism, who tend to experience more negative emotions (McCrae & Costa, 1987), are likely to experience lower job satisfaction. In a meta-analysis study, Judge, Heller, and Mount (2002, p. 530) found that 'the relations of neuroticism and extraversion with job satisfaction generalized across studies'. Apart from affecting the tonic level of well-being, these traits are also associated with differential emotional sensitivity to positive and negative life events. The biopsychological theory of personality (Gray, 1981(Gray, , 1987 argues that emotions and behaviours are neurologically regulated by two motivation systems: the behaviour activation system (BAS) and the behaviour inhibition system (BIS). BAS is associated with increased sensitivity to signals of reward, and BIS is associated with increased sensitivity to signals of punishment. People high in extraversion have a stronger operation of the BAS system and react more strongly to rewards and pleasure. In contrast, people high in neuroticism have a stronger operation of the BIS system and react more strongly to punishment and losses (Elliot & Thrash, 2002).
Experimental research has produced support for this view by showing that individuals often selectively attend to, retrieve and process information in ways that are congruent with their underlying personality traits. Bower and Cohen (1982) argue that emotions influence the way the brain stores and organises information and memories. Individuals with heightened susceptibility to positive emotions tend to notice and retrieve positive information with greater ease because of the intimate connection of emotions and cognitions in their neural networks. Exposing subjects to a variety of images, Larsen andKetelaar (1989, 1991) found that extraversion was correlated with elevated positive affect following the viewing of positive images while neuroticism was correlated with elevated negative affect following the viewing of negative images. Reed and Derryberry (1995) used response time as a measure of the speed of discriminating various types of emotional stimuli and found that extraversion was associated with the faster detection of positively valenced words and neuroticism was associated with the faster detection of negatively valenced words. Similar findings of trait-congruency in information processing have also been found with respect to information retrieval (Lishman, 1974;Mayo, 1989) and interpretation of life events (MacLeod & Cohen, 1993;Richards, Reynolds, & French, 1993).
In the context of the present study, a career transition can either represent a 'gain' or a 'loss' depending on the direction of occupational mobility. Based on the biopsychological theory of personality, individuals with different levels of extraversion and neuroticism are differentially susceptible to positive and negative events because of differences in their BAS and BIS systems. Specifically, individuals high in extraversion will experience a greater surge of subjective well-being in the condition of upward occupational mobility because of their tendency to focus attention on reward stimuli. For instance, they are more likely to notice the positive consequences of upward career mobility such as increased development opportunities and enlarged social network. By contrast, individuals high in neuroticism are likely to suffer a greater decline of subjective well-being in the condition of downward occupational mobility because of their raised sensitivity to threatening stimuli. Compared with their emotionally stable counterparts, highly neurotic people are quicker to notice the negative consequences of downward career mobility such as the loss of valued job features and decline in occupational prestige, which can lead to stronger feelings of deprivation and frustration. Based on the discussion, we propose Hypothesis 5. Individuals high in extraversion experience a greater increase in job satisfaction upon upward occupational mobility compared with those low in extraversion.
Hypothesis 6. Individuals high in neuroticism experience a greater decrease in job satisfaction upon downward occupational mobility compared with those low in neuroticism.

| Individual differences in adaptation to occupational mobility
Adaptation is a process that allows constant stimuli to fade into the background so that individuals can free up mental resources to deal with new stimuli in the environment that request immediate attention and actions (Fredrick & Loewenstein, 1999). Accordingly, the In this study, we propose that individuals high in extraversion and neuroticism will experience stronger initial reactions to upward and downward occupational mobility, respectively. Given the intimate link of reaction and adaptation, we expect that these people will experience longer periods of adaptation following the career transition. Due to their raised sensitivity to gains, extroverts will experience greater feelings of joy upon moving up the career ladder. They are quicker to notice the pleasant aspects of their surroundings and slower to shift their attention away from pleasant thoughts. By contrast, individuals high in neuroticism will experience more intense feelings of disappointment following downward career mobility because of their elevated sensitivity to losses. They may ruminate on the negative consequences of downward career transitions and adopt ineffective coping strategies such as regretting and self-blaming (Gunthert, Cohen, & Armeli, 1999;Nolan, Roberts, & Gotlib, 1998), which can lead to prolonged frustration and despondency. Based on the discussion, we derived our final hypotheses: Hypothesis 7. Individuals high in extraversion take longer time to return to their baseline job satisfaction following upward occupational mobility compared to those low in extraversion.
Hypothesis 8. Individuals high in neuroticism take longer time to return to their baseline job satisfaction following downward occupational mobility compared to those low in neuroticism.

| Job satisfaction
In the BHPS, job satisfaction is measured by a simple question that asked respondents to rate on a seven-point scale how satisfied they are with their present jobs. Research shows that the single-item measure of job satisfaction has acceptable reliability compared with composite measures derived from multiple items (Wanous, Reichers, & Hudy, 1997). The measure has been widely used in previous research (e.g., Chaudhuri, Reilly, & Spencer, 2015;Georgellis & Tabvuma, 2010;Zhou et al., 2017).

| Occupational mobility
Occupational mobility is measured by comparing an individual's occupational code before and after turnover based on the Standard Occupational Classification 1990 (SOC90) system used by the U.K. Office for National Statistics. Following the occupational mobility literature (Hadjar & Samuel, 2015;Markey & Parks, 1989), we use change of major occupational groups to measure the direction of occupational mobility. A potential caveat of this approach is that occupational class categories are not strictly hierarchical and the problem is especially salient among the intermediate occupational positions (Hadjar & Samuel, 2015). For instance, it can be difficult to determine whether a change of job from clerical to craft or service work represents upward or downward mobility. Given this concern, we adopt a relatively conservative approach in which managers, professionals, and associate professionals (1-digit SOC90 codes: 1, 2 and 3) are grouped into a highly skilled occupational class category; clerical, craft and personal service occupations (1-digit SOC90 codes: 4, 5 and 6) are grouped into an intermediate occupational class category, while sales, machine operatives and elementary occupations (1-digit SOC90 codes: 7, 8 and 9) are grouped into a low-skilled occupational class category, since there is broad consensus in the literature that these categories generally capture jobs with high, medium and low skill requirements (e.g., Inanc, Zhou, Gallie, Felstead, & Green, 2015;McGinnity & Russell, 2013).  (Table 1). We have carried out a range of robustness checks to ensure the validity of our measure of occupational mobility. 2

| Extraversion and neuroticism
In 2005, the BHPS introduced a set of questions on personality based on Benet-Martínez and John (1998)'s measures of the Big Five personality dimensions. Extraversion is measured by three questions that asked individuals the extent to which they saw themselves as someone who is 'outgoing, sociable', 'reserved' and 'talkative', and neuroticism is measured by three questions that asked the extent to which they saw themselves as someone who 'worries a lot', 'gets nervous easily' and 'is relaxed, handles stress well'. Answers were made on a 7-point scale running from 'Does not apply to me at all' to 'Applies to me perfectly'. The Cronbach's alpha coefficient is .75 for the extraversion scale and .68 for the neuroticism scale. We have used respondents' unique personal id to match their personality scores across waves so that the same person is given the same extraversion and neuroticism scores across the whole survey period. 3 To take into account measurement errors, we have carried out confirmatory factor analyses to calculate latent factor scores of extraversion and neuroticism and used these scores in our analysis.

| Analytical procedures
We have used fixed effect models to analyse intra-individual job satisfaction trajectory following each type of occupational mobility. The key advantage of fixed effect modelling lies in its ability to filter out unobserved individual heterogeneity that may confound the impact of the predictor variables. By focusing on intra-individual variations over time, fixed effect models remove the omitted variable bias which often plagues the validity of causal inferences (Allison, 2009;McNeish & Kelley, 2018;Wooldridge, 2010). To capture the timing of job change, we followed Clark, Diener, Georgellis, and Lucas (2008) to create lead and lag dummies to measure job satisfaction trajectory during the turnover process. Lead dummies identify up to 4 years before turnover and lag dummies identify up to 4 years after turnover.
For example, lead 0-1 indicates a year before turnover, lag 0-1 indicates the year in which turnover occurs, lag 1-2 indicates a year after turnover, and so on. If someone experiences downward occupational mobility at t and upward occupational mobility at t + 3, he or she will be followed for 2 years after the first downward occupational mobility, while the second transition made at t + 3 will be treated as a new start (t) for upward occupational mobility and followed for as long as the person stays in the same occupation. The same respondent can thus contribute to upward, lateral, and downward occupational mobility observations depending on his or her career trajectory. The numbers of leads and lags for each type of occupational mobility are shown in Table 2.
We first estimate the effects of upward, lateral and downward occupational mobility on job satisfaction trajectory for all employees and then interact the lead and lag dummies with extraversion and neuroticism to examine whether the temporal effects of occupational mobility vary across individuals with different personality. The moderation analyses were carried out on the full sample. To illustrate 2 First, we compared a range of job quality indicators across occupational categories based on information from the U.K. Skills and Employment Surveys (SES) and found employees in higher occupational classes generally report higher extrinsic and intrinsic job rewards than those in lower occupational class positions. Second, we compared the well-being impact of long-range and short-range occupational mobility and found the former has greater effects on job satisfaction than the latter. Finally, we used the International Standard Classification of Occupations 1988 (ISCO88) to construct our occupational measures and found the same pattern of results as those based on SOC90. 3 We acknowledge this approach relies on the assumption that personality does not vary over time. Although research shows that personality is generally stable over the life course (e.g., Roberts & DelVecchio, 2000), counter-argument also exists (e.g., Roberts & Mroczek, 2008). Therefore, we have performed moderation analysis not only by using individual personality scores but also divided participants into different groups according to their ranks on the personality scale (e.g., high vs. low in neuroticism). As the rank order of personality scores are relatively stable in adulthood (Roberts & DelVecchio, 2000), such analysis is less susceptible to changes in individuals' personality scores over time. We have also experimented with different cut-off thresholds when dividing the sample (e.g., top fifth vs. bottom fifth, top third vs. bottom third; top half vs. bottom half) and obtained consistent results that show higher levels of neuroticism amplify the negative effect of downward occupational mobility.
these results visually, we selected two groups of individuals with high and low personality scores and graphed their job satisfaction trajectories during each type of occupational mobility. In all models, we control for time-varying individual and workplace characteristics in line with previous research (e.g., Chadi & Hetschko, 2018;Clark, Diener, Georgellis, & Lucas, 2008;Georgellis & Tabvuma, 2010 Four percent of job satisfaction data are missing from the analytical sample. With the assumption that data were missing at random, we have applied multiple imputation procedure (using the MI commands in Stata 15) to impute missing information on job satisfaction based on individuals' demographic characteristics and conducted the fixed effect analyses on the imputed data. 5

| RESULTS
We start by examining the overall effects of upward, lateral and downward occupational mobility on job satisfaction for all employees. The main independent variables of interest are the lead and lag dummies that capture the temporal change in job satisfaction before, during and after each type of occupational change (see Table 3 and Figure 1a). The baseline is denoted by '0' in Figure 1a, which is measured by the average job satisfaction score reported by an individual over the entire survey period. A significant and positive coefficient for a time dummy indicates that job satisfaction is above the baseline in the reference year, while a significant and negative coefficient indicates the opposite. We have performed the analysis without and with controls. As reported in Table 3, including control variables or not did not affect our conclusion.  Note: Leads measure the number of years before occupational change and lags measure the number of years after occupational change. For instance, lag 0-1 year indicates the year in which the transition occurs, lag 1-2 years indicates a year after the transition, lag 2-3 years indicates 2 years after the transition, and so forth.
effect' at the time of turnover, while downward occupational mobility has the opposite effect. The pattern of adaptation shows an interesting asymmetry between upward and downward mobility.
While individuals who moved up the occupational ladder returned to their baseline well-being by the third year after turnover, those who moved down the occupational ladder suffered a prolonged loss of well-being that continued for at least 4 years after the transition.
These results are consistent with Hypothesis 3 but inconsistent with Hypothesis 4.
Next, we carried out moderation analysis by interacting occupational mobility lag dummies with extraversion and neuroticism scores to examine whether the pattern of job satisfaction trajectory differs across individuals with different personality traits. 6 Table 4 shows 6 Although this study was focused on extraversion and neuroticism, we have also analysed the effect of openness, agreeableness and conscientiousness to provide a complete picture of the role of Big Five personality traits in individuals' reactions to career changes. Our analyses showed that these traits did not significantly moderate the effect of occupational mobility on post-turnover job satisfaction trajectory. These results can be found in Table A1. 7 Results of analyses without controls are available upon request.  there is no significant interaction effect between extraversion and the lag dummies for upward occupational mobility, which does not support Hypotheses 5 and 7. Neuroticism does not have a significant interaction effect with the lag dummy for downward occupational mobility in the year of turnover, failing to support Hypothesis 6. Neuroticism, however, has significant and negative interaction effects with the lag dummies in the third and fourth years after turnover, suggesting that individuals high in neuroticism experience a worsening trend in job satisfaction following downward occupational mobility.
This evidence does not allow us to make a definitive conclusion for Hypothesis 8 because longer-running panel data are needed to examine whether, and if so, how long it will take for individuals high in neuroticism to return to their baseline.
To illustrate the moderation effect of neuroticism, we have plotted the results from the fixed effect models in Figure 1b   mobility. The pattern may reflect an entrapment effect (Taylor, Gooding, Wood, & Tarrier, 2011) such that individuals become depressed when they are stuck in undesirable situations. We did not have sufficient evidence to conclude on whether individuals experiencing downward occupational mobility will eventually return to their baseline well-being as such analysis requires data with post-turnover observations over longer time. This study has also examined individual differences in reactions and adaptations to different types of career change. Contradicting our initial expectation, extroverts did not react more favourably to upward occupational mobility than introverts. We speculate that extroverts, with higher tonic levels of well-being (Lucas & Baird, 2004), may need stronger positive stimuli to further boost their job satisfaction above the baseline. This speculation is in line with prospect theory (Kahneman & Tversky, 1979) that suggests that compared with losses, a greater amount of gains is needed to trigger the same amount of change in psychological well-being. This proposition may be examined in future studies. In contrast, we found that individuals high in neuroticism experienced a steeper decline of job satisfaction in the condition of downward occupational mobility. The negative effect, however, did not emerge at the time when they moved down the occupational ladder but in the subsequent years. This finding is intriguing because it shows that those high in neuroticism only became more miserable after having spent some time in their new jobs.
Our findings have important implications for personality research.
While the effect of personality on employee outcomes has been extensively studied, they are more often investigated at the between- Finally, our study contributes to the hedonic adaptation literature.
Set point theory suggests that individuals will return to their baseline subjective well-being after the influences of life events fade away.
Empirical studies have been conducted to test the idea against different life events and the extant literature suggests that people tend to adapt to positive events quickly, such as marriage (Lucas et al., 2003;Lucas & Clark, 2006) and voluntary job change (Boswell et al., 2005), but not to negative events such as unemployment  and widowhood (Lucas et al., 2003). Similarly, we find that it is more difficult for people to adapt to downward than to upward occupational mobility. The fact that upward and downward career change have asymmetrical effects on job satisfaction supports Kahneman and Tversky's (1979) argument that losses have a greater impact than gains on subjective well-being. In other words, 'bad is stronger than good' (Baumeister, Bratslavsky, Finkenauer, & Vohs, 2001;Taylor, 1991). We join Diener et al., (2006)  Additionally, this study highlights the need to pay special attention to employees who have moved from higher to lower skilled occupations as this group is vulnerable to a long-term decline of job satisfaction following the transition. The risk is particularly high among those who are characterised by high levels of neuroticism.
Although job quality is partially determined by occupation (William, Zhou, & Zou, 2020;Williams, 2017a;Williams, 2017b;Zou, 2015), prior research shows that enriched job design that allows sufficient discretion over task planning and execution has significantly positive effects on employee well-being (Gallie & Zhou, 2020;Gallie, Zhou, Felstead, Green, & Henseke, 2017;Gallie, Zhou, Felstead, & Green, 2012;Wu, Griffin, & Parker, 2015). We speculate these benefits are more pronounced for those who are overqualified for their jobs, which often results from downward occupational mobility. Human resource practices that support autonomy and job crafting can potentially mitigate the negative impact of downward occupational mobility by helping individuals find new ways to engage their talents and skills (Wrzesniewski & Dutton, 2001;.

| Limitations
This study has a few limitations. First, we do not have information on the reason for occupational mobility since it was measured by comparing an individual's Standard Occupational Classification codes across adjacent years. While acknowledging that reasons for job change may influence post-turnover well-being, we believe our findings cannot be solely explained by the motive of job change. Voluntary job changes can lead to upward occupational mobility, but the link is by no means universal. For instance, while previous research shows that most job changes are voluntary (e.g., Markey & Parks, 1989, 1989Polsky, 1999), we find that only a quarter of career For instance, Georgellis, Lange, and Tabvuma (2012) find that job satisfaction score declines by .3 to .4 on a 7-point scale among female workers following the birth of the first child, most likely due to the rise of work life conflicts. In contrast, this study shows that job satisfaction score declines by .6 on the same 7-point scale among those high in neuroticism 4 years following downward occupational mobility, which represents a significant loss of well-being in the context of work-related life events.

| CONCLUSION
This study has adopted an integrated perspective that combines situational and dispositional perspectives to advance our understanding of the long-term impact of career changes on subjective well-being. In   His research specialism is mapping socio-economic disparities in the labour market. Much of his work has focused on pay disparities across the occupational structure. In more recent years, he has developed interests in the quality of work more broadly (job quality, wellbeing).