Previous studies have paid little attention to the employees’ ability to exit a job-lock situation and factors that determine this ability. It remains unclear why some employees who experience job lock are able to exit this state while others remain in job lock. We use longitudinal data to identify employees who have fallen in the state of job lock and their subsequent behavior—exiting or remaining in job lock. By use of a first-order Markov transition models, we analyze the relevance of sociodemographic features, employment, occupational, sectoral, and contextual factors, as well as personality characteristics in explaining the transition or its absence. Overall the results show that both demographic factors and work-related aspects increase the likelihood that an employee enters the long-term job lock state (especially for older, married, full-time employed, those in a craft occupation and governmental sector, and in a region with high unemployment). Mental health problems and personality characteristics (low peak-end self-esteem and decisional procrastination) have a significant effect on the probability to stay in long-term job lock. On the contrary, having a managerial, service, or associate occupation, working in the private sector, and having promotion opportunities increase the chance of an exit from the state of job lock.
The desire to adapt to feelings of dissatisfaction is natural. Dissatisfied employees are likely to try to reduce their job dissatisfaction and work-related stress by adjusting to their current job or by changing jobs [
Various studies in the fields of economics and psychology have investigated the phenomenon of job lock and its determinants following the perspective of their own field (some examples include [
Nevertheless, both economics and psychology studies have paid little attention to the employees’ ability to exit a job-lock situation and factors that determine this ability. It remains unclear why some employees who experience job lock are able to exit this state while others remain in job lock. Also, for those employees who leave the state of job lock, it may be asked what the mechanisms are by which this happens—do they adjust by becoming satisfied, do they use mobility as a way of dealing with dissatisfaction, or a combination of both? The answers to these questions are important in developing interventions to assist employees to reduce work stress and successfully adapt to job dissatisfaction.
The aim of this paper is to investigate the process of transition from a job-lock situation (i.e., being dissatisfied with the job but remaining in the same job) to other states, for example, adjusting and becoming satisfied in the same job (immobile and job satisfied), changing jobs and becoming satisfied (mobile and job satisfied), or changing jobs but again become dissatisfied with the new job (mobile and job dissatisfied).
We compare those in job lock, who fail to make a transition, to those who experience one of the three transitions described previously. Also, we compare the transition processes
We combine insights from both economics and psychology studies to identify a set of possible transition determinants. In particular, based on Huysse-Gaytandjieva et al. [
We use data from the British Household Panel Survey [
Section
Review studies in the area of economics and psychology [
Successful adaptation to job dissatisfaction is seen as an alleviation of the job dissatisfaction level as a result of engaging in some adjusting mechanism [
There are various personality characteristics related to responses to dissatisfaction and adaptation. The value of self-esteem in the adaptation process is in particular emphasized in the literature [
Self-esteem is shown to be a personality characteristic that protects people against stressful consequences [
Much of the research about the relationship between self-esteem and health appears to have been done in terms of the influence of self-esteem on health-related behaviors. From the other side, in a review of the self-esteem literature, Baumeister et al. [
Furthermore, a negative self-image is important for the occurrence of procrastination [
Procrastination may become dysfunctional when people frequently habitually delay to begin or complete tasks [
Given the aforementioned, additionally to self-esteem, in this study, we include procrastination as a variable that can be seen as a consequence of preexisting personality characteristics as self-esteem and as “an agent for bringing about adverse consequences of its own right” [
We use data from the British Household Panel Survey (BHPS). The BHPS is an annual longitudinal survey based on a nationally representative sample of about 10,000 adults in Great Britain. Individuals are interviewed in successive waves. Details about the survey can be found in Taylor et al. [
To construct the models for our analysis, we use data related to job dissatisfaction and job immobility provided by the BHPS dataset. In particular, the job dissatisfaction variable for our analysis is derived from the BHPS variable that indicates the overall job satisfaction of a respondent measured on a seven-point Likert scale. Thus, we construct a dummy job-dissatisfaction variable for each year (0 = job satisfaction; 1 = job dissatisfaction). The category “neither satisfied nor dissatisfied” is seen as indicative of not being all that satisfied with the job [
We derive the job immobility variable from the BHPS variable that indicates tenure: “What was the date you started working in your present position, by that I mean the beginning of your current spell of the job you are doing now for your present employer?”. If in a given year, tenure is greater or equal to one year, job immobility is coded with zero, and if tenure is less than one year, job immobility is coded with one. Thus, a dummy immobility variable is constructed for each year.
We use the operationalization of job lock provided by Huysse-Gaytandjieva et al. [
The job-dissatisfaction and job-immobility variables described previously, as well as the operational definition of job lock, are used to construct two nominal dependent variables for our analysis to present transitions to and from a job-lock state, respectively, transitions transitions
The two transition models are schematically presented in Figure
Transition models.
We also define one binary dependent variable to compare those in a job-lock state (dissatisfied and immobile for two subsequent years, coded with 1) to those who are dissatisfied with their job but remain mobile during at least one of the years (not in a job lock situation even though job dissatisfied for two subsequent years, coded with 0). This way we include in our analysis all employees, who reported job dissatisfaction for two subsequent years (prolonged job dissatisfaction). The rest of the employees are omitted from the analysis.
The explanatory variables for our analysis represent six groups of factors that previous (economics and psychology) studies indicate as relevant in analyzing the state of job lock or its absence [
We use the response to the following question as an indicator of self-esteem: “Have you recently been thinking of yourself as a worthless person?” (0 = high, stable self-esteem; 1 = unstable, low self-esteem). The question is taken from the General Health Questionnaire (GHQ) included in the BHPS. The GHQ has been validated in nine countries [
None of the existing measures of procrastination are directly applicable to work-related behavior [
Health problems related to anxiety, depression, and so forth (“Do you have any of the health problems or disabilities: anxiety, depression or bad nerves, psychiatric problems”) are constructed as a dummy variable (coded: 0 = absence; 1 = presence). Age is measured as a continuous variable. The variables gender, marital status, working full time, member of the trade union, opportunities for promotion in the current job, belonging to the employer’s pension scheme, and training as a part of the present employment are included in the analyses as dummies. Occupation is measured by the standard occupational classification (SOC). Nine dummy variables are included for occupation. Further, type of sector is included in the analysis as four dummy variables.
For an easier interpretation of the regression results, Table
Bivariate correlation analysis provides supporting information to design the model [
Table
Descriptive statistics.
Job dissatisfaction during two subsequent years and |
Frequency | Transition from |
Frequency | Transition from job lock to |
Frequency | |||
---|---|---|---|---|---|---|---|---|
|
% |
|
% |
|
% | |||
0 = mobile during at least one of the years (not in job lock) | 1605 | 54.4 | 1 = dissatisfied and immobile | 484 | 61.4 | 1 = dissatisfied and immobile | 364 | 44.1 |
2 = dissatisfied and mobile | 51 | 6.5 | 2 = dissatisfied and mobile | 78 | 9.5 | |||
| ||||||||
1 = immobile during both years (in job lock) | 1344 | 45.6 | 3 = satisfied and immobile | 206 | 26.1 | 3 = satisfied and immobile | 267 | 32.4 |
4 = satisfied and mobile | 47 | 6.0 | 4 = satisfied and mobile | 116 | 14.1 | |||
| ||||||||
Total | 2949 | 100 | 788 | 100 | 825 | 100 |
Results of binary probit regression.
Explanatory variable |
Job dissatisfaction for two subsequent years | ||
---|---|---|---|
0 = not in job lock (i.e., mobile during at least one of the years) | |||
1 = in job lock (i.e., immobile during both years) | |||
Coefficient | Std. error | Odds ratio | |
Sociodemographic features | |||
Age | 0.024** | 0.003 | 1.035 |
Gender | 0.139* | 0.058 | 1.261 |
Health status |
|
0.106 | 0.505 |
Marital status | 0.127* | 0.057 | 1.234 |
Personality attributes: | |||
Min peak-end self-esteem | 0.054 | 0.073 | 1.076 |
Procrastination | −0.049 | 0.084 | 0.932 |
Type of occupation (reference category: clerical and secretarial) | |||
Manager and administrators | −0.369** | 0.097 | 0.545 |
Professional | 0.030 | 0.100 | 0.946 |
Associate professional/technical | −0.084 | 0.105 | 0.875 |
Craft | 0.448** | 0.094 | 2.093 |
Personal and protective service | −0.203 | 0.108 | 0.718 |
Sales | −0.064 | 0.116 | 0.891 |
Plant and machine | 0.095 | 0.088 | 1.171 |
Other occupations | 0.079 | 0.110 | 1.167 |
Employment conditions | |||
Full-time contract | 0.360** | 0.087 | 1.794 |
Employer pension scheme | 0.318** | 0.076 | 1.677 |
Member of the trade unions | −0.035 | 0.082 | 0.949 |
On-the-job training | 0.071 | 0.055 | 1.128 |
Promotion opportunities | −0.627** | 0.068 | 0.355 |
Type of sector (reference category: army and other sectors) | |||
Civil | −0.388** | 0.141 | 0.524 |
Governmental | −0.580** | 0.115 | 0.382 |
NHS or higher education | −0.867** | 0.142 | 0.236 |
National industry | 0.099 | 0.207 | 1.137 |
Nonprofit | −0.859** | 0.213 | 0.239 |
Private | −0.928** | 0.085 | 0.214 |
Work-related contextual features | |||
Regional unemployment rate |
|
0.013 | 0.995 |
Intercept |
|
0.220 | |
Observations | 2949 | ||
Pseudo |
0.147 |
Of these, 1344 respondents experience job lock (i.e., they remain immobile during the two years) and 1605 respondents are mobile during at least one of the years. As the regression results suggest (see Table
As much as 61.4% of those who experience job lock (see Table
Table
Results of multinominal logistic regression.
Explanatory variables |
Transition from job lock to |
Transition from |
||||
---|---|---|---|---|---|---|
Dissatisfied and mobile | Satisfied and immobile | Satisfied and mobile | Dissatisfied and mobile | Satisfied and immobile | Satisfied and mobile | |
Odds ratio | Odds ratio | Odds ratio | Odds ratio | Odds ratio | Odds ratio | |
Sociodemographic features | ||||||
Age | 0.977 | 0.981* | 0.933** | 1.008 | 0.989 | 0.928** |
Gender | 1.298 | 0.698 | 0.577 | 0.751 | 0.764 | 0.662 |
Marital status | 0.600 | 0.905 | 0.846 | 0.405** | 0.743 | 0.496 |
Health problems: anxiety, depression, and so forth | 2.129 | 0.814 | 1.457 | 0.499 | 0.941 | 0.447 |
| ||||||
Personality attributes: | ||||||
Min peak-end self-esteem | 0.689 | 0.409** | 0.413* | 0.685 | 0.557* | 0.121** |
Procrastination | 1.137 | 1.149 | 1.436 | 0.609 | 1.982* | 3.259* |
| ||||||
Type of occupation (reference category: clerical and secretarial): | ||||||
Manager and administrators | 2.974* | 2.285* | 2.185 | 1.976 | 3.371** | 12.505** |
Professional | 0.870 | 1.841 | 2.109 | 2.404 | 1.686 | 8.758** |
Associate professional/technical | 2.713 | 3.805** | 2.652 | 0.473 | 2.989** | 7.848* |
Craft | 0.342 | 1.521 | 0.913 | 0.100* | 1.837* | 3.725 |
Personal and protective service | 1.563 | 3.381** | 4.110* | 7.748** | 3.361** | 3.215 |
Sales | 1.820 | 0.905 | 1.947 | 1.445 | 1.560 | 7.070* |
Plant and machine | 0.296 | 0.925 | 0.396 | 0.778 | 1.171 | 6.499** |
Other occupations | 0.706 | 1.465 | 0.580 | 2.086 | 1.712 | 1.903 |
| ||||||
Type of sector: (reference category: other sectors): | ||||||
Private | 1.899 | 1.977* | 3.112** | 3.909* | 1.114 | 0.225** |
Civil | 1.731 | 0.999 | 1.145 | 2.002 | 0.631 | 1.394 |
Governmental | 1.239 | 1.346 | 0.907 | 0.847 | 1.293 | 0.276* |
| ||||||
Employment conditions: | ||||||
Full-time contract | 0.430 | 0.627 | 0.129** | 1.447 | 0.735 | 0.386 |
Employer pension scheme | 1.796 | 1.349 | 1.811 | 0.832 | 1.508 | 0.660 |
On-job training | 0.984 | 1.132 | 0.764 | 1.414 | 0.923 | 0.685 |
Promotion opportunities | 3.991** | 2.329** | 14.572** | 0.829 | 0.985 | 0.976 |
Member of the trade unions | 0.532 | 0.667 | 0.608 | 0.546 | 0.968 | 1.636 |
| ||||||
Work-related contextual features: | ||||||
Regional unemployment rate | 0.836** | 0.864** | 0.818** | 0.857* | 0.794** | 0.828* |
| ||||||
Other model characteristics | Observations = 825 | Observations = 782 | ||||
LR |
LR |
|||||
Pseudo |
Pseudo |
Interaction terms.
Variables/interactions | Satisfied and mobile |
---|---|
Dependent variable: transition from job lock to |
|
| |
Peak-end self-esteem |
0.076* |
Peak-end self-esteem | 3.944 |
Mental health problems | 3.114 |
| |
Mental health problems |
0.088* |
Mental health problems | 2.643 |
Procrastination | 10.680* |
| |
Age |
0.897* |
Peak-end self-esteem | 19.341 |
Age | 0.942** |
| |
Regional unemployment rate |
1.461* |
Peak-end self-esteem | 0.014** |
Regional unemployment rate | 0.774** |
| |
Satisfied and immobile | |
| |
Dependent variable: |
|
| |
Regional unemployment rate |
1.361* |
Peak-end self-esteem | 0.033** |
Regional unemployment rate | 0.759** |
Coding of the dummy variables used in the analysis.
Dummy variables | Dummy codes | |
---|---|---|
0 | 1 | |
Gender | Female | Male |
Marital status | Separated; divorced; widowed; never married | Married |
Health problems: anxiety, depression, and so forth | No | Anxiety, depression, or bad nerves |
Decisional procrastination | More than usual; same as usual | Less so; much less |
Self-worth |
Not at all; no more than usual |
Rather more; much more |
Full-time contract |
No |
Yes |
As indicated in Table
As suggested by Table
Four interactions show a significant effect. First is the interaction between peak-end self-esteem and health for the transition from “satisfied and mobile.” At the same time the effect of peak-end self-esteem variable becomes not significant while the effect of the health variable does not change. Second is the interaction between procrastination and health. This changes the effect of procrastination to become significant while keeping the effect of the health variable. Third is the interaction between age and peak-end self-esteem. This changes the effect of self-esteem to insignificant while keeping the effect of age. Fourth is the interaction between peak-end self-esteem and the regional unemployment rate.
Additionally, we checked how many employees stay dissatisfied and immobile for 4 and 5 subsequent years. They are 136 and 47, respectively, which shows a decreasing trend.
Overall the results show that being older, being married, working in a craft occupation, in the governmental sector, having a full-time job and high regional unemployment rate increase the likelihood that an employee enters the long-term job lock state. Furthermore, low peak-end self-esteem, mental health problems, and decisional procrastination show significant effects on the probability to stay in long-term job lock (failure to exit the job-lock state). On the contrary, having a managerial, service, or associate occupation, working in the private sector, and having promotion opportunities increase the chance of an exit from the state of job lock. A high regional unemployment rate is not statistically significant for those dissatisfied in two consequent years. It seems that a high regional unemployment rate provides incentives for employees who are dissatisfied to adapt by adjusting. Among dissatisfied employees, older workers are less likely to use mobility as an adaptation strategy. Further, possessing a company pension scheme increases the likelihood that the employee, who is dissatisfied, is immobile for two subsequent years.
Our results highlight the process of the transition to and from a job-lock situation, as well as the situation of long-term job lock. We briefly discuss the key findings in the subsequent paragraphs.
Our results suggest that being older, married, with low peak-end self-esteem, working in a craft occupation, in the governmental sector, and high regional unemployment rate are push factors to a job-lock state.
As previous research also shows, elderly employees are less mobile [
Another important variable concerning mobility is marital status. Being married is negatively correlated with the probability of quitting when dissatisfied with the job [
Employment, occupational, sectoral, and contextual factors also may push individuals to a job-lock state. As previous studies have shown when the regional unemployment rate increases, employees are more likely to be immobile [
The factors that push an employee into job lock—age, low peak-end self-esteem, and high regional unemployment rate—also play an essential role in the failure to exit job lock and enter long-term job lock. In addition to the already discussed variables, having a full-time contract increases the probability to enter the long-term “job lock” state rather than to move to any of the other states. This confirms the outcome of previous studies that full-time workers are less mobile [
Some occupations play a significant role in the transitions from “job-lock” state. In particular, being in a managerial or administrator position, associate professional, personal and protective service increases the chance that the employee exits the “job-lock” state. Further, private sector and promotion opportunities pull employees out of job lock. Holding a personal and protective occupation increases the chance that an employee moves to “satisfied and immobile” or “satisfied and mobile” instead of remaining in a job-lock state.
Individual abilities play an important role in occupational decision making [
Besides, having promotion opportunities in the current job increases the likelihood that the employee moves to one of the other three states. In general having promotion opportunities in the current job increases overall job satisfaction [
Employees in a manager and administrator occupation, those in personal and protective service occupations, with promotion opportunities, working in the private sector use much more often mobility as a coping strategy in order to adjust to job dissatisfaction.
Thus, managers and administrators and personal and protective service occupations are capable of successfully adapting to job dissatisfaction by using active forms of adaptation. It might be that for employees in those occupations, job dissatisfaction is just one of the drivers for job mobility. Qualities to succeed in your job may be also essential qualities for successful adaptation.
At the same time, manager and administrator occupation, associate professional, personal and protective service occupations, sales, working in the private sector, and with promotion opportunities, employ work adjustments as coping strategy (satisfied and immobile).
Thus, almost the same variables play a role in successful adaptation independent from the form of the adaptive strategy (either with job satisfaction or job mobility). We can conclude that people are either capable of successfully adapting or not irrespective of the coping strategy used.
Peak-end low self-esteem for both multinominal models is related to the transition to job lock or long-term “job lock.” On the other hand, the analysis of employees who are dissatisfied for two years does not show self-esteem to be significant compared to those who moved and those who stay in a job lock. Further, the findings support our expectation that the inclusion of the procrastination and psychological health variables in our model leads to a better explanation of long-term “job lock.” Peak-end low self-esteem, mental health problems, and decisional procrastination show significant effects on the probability to stay in long-term “job lock.” Low self-esteem plays a role in failure to adapt to job dissatisfaction. However, having high self-esteem does not show guarantee for successful adaptation.
Low self-esteem and mental health problems have a joint effect on the transition from a job-lock state (shown by the significance of the interaction term). This is in line with the results from previous research that depression and anxiety are some of the symptoms experienced by people with low self-esteem [
Additionally, the analysis comparing coping strategies by dissatisfied employees shows that those with poor health are much more often mobile. It might be that health deterioration leads to an adjusted job, internal mobility, or what some other studies show—the severer the mental health the sooner the employee leaves [
Due to the fact that procrastination has been seen as a risk factor for more serious depression and anxiety [
We find that when regional unemployment rate is high, employees with low self-esteem have less chance to experience job dissatisfaction. It might be that when there are fewer opportunities on the labor market, people see their present job in a more favorable light and report more satisfaction. Employees may also realize that they are happy to have a job, or it could be a selection effect where dissatisfied employees are more likely to be laid off. Additionally, this employees’ behavior might be influenced not by the actual job availability but by their perceptions of job availability [
The aim of this study has been to explore the process of the transition to and from a job-lock situation, as well as the situation of long-term job lock. Our results provide insights into understanding individual differences in adaptation which help to illuminate when and why successful adaptation does or does not occur.
Another contribution of the present study is that by following the process—from dissatisfaction to job lock to long-term job lock—we are able to distinguish the essential variables which play a role in the state transitions. Further, we faced Diener et al.’s [
Nevertheless, the current findings should be interpreted with caution because of the following limitations of the present study: lack of differentiation of voluntary and involuntary job mobility, mental health problems, and the usage of a proxy measure of self-esteem and procrastination. Nonetheless, none of the existing measures of trait procrastination are directly appropriate to work-related behavior [
Irrespective of the study limitations, the results can be applied on different levels: at the individual, organizational (HRM, company doctors, managers, coaches, mental health professionals), and societal levels (labor unions, government) by increasing awareness and knowledge, using them for prevention, in problem solving, and development of supportive programs.
In terms of practical implications, the current findings present a need to develop and incorporate programs tackling task avoidance and procrastination at the work place. Additionally, counseling to support employees with dysfunctional procrastination tendency can be beneficial as employee’s stress reduction may increase productivity. The results might be beneficial to those who are involved in employee selection and those who are responsible for making promotion decisions.