changeImage by geralt, under a CC0 licence

Searching and hiring workers imposes high costs on firms, not least when it comes to vacancies that need to be filled with extraordinary talents. Once a worker is hired, it is important to ensure a successful integration of the newcomer. The key process here is organisational socialisation, which renders a new worker, who is an outsider at the beginning, an insider. It is up to the firm to shape this process successfully, i.e. to motivate employees to remain in the firm.

Prominent examples are game rooms helping employees to socialise and feel at home as well as the German trend to appoint so-called ‘feel-good managers’ who care about workers’ well-being in general and about newcomers’ needs in particular. Probably not by chance such trends often stem from the tech start-up industry where turnover is high and talents are in great demand.

Job satisfaction is regarded a key to hindering workers from leaving the firm. It is therefore important to study at which points in time after a job change newcomers are relatively satisfied, and when they are irked and become thus more likely to leave. Our analyses rely on a large dataset being representative for the German population. The ‘Socio-Economic Panel’ study (‘SOEP’) interviews the same people repeatedly, which allows us to observe workers throughout their careers and when they change from one job to another.

Thanks to a multi-faceted questionnaire, we can distinguish between different types of job changes, namely intended moves triggered by workers resigning and unintended job changes following plant closures. Whereas the former typically have good reasons to switch jobs and are therefore expected to be relatively satisfied afterwards, the latter case informs us about the isolated impact of switching workplaces when there is no particular intention to do so.

Changing employers can change one’s entire life, the place and even the country of residence. One particular area that might be affected indirectly, for example because of higher work hours, is family life. Hence, we argue that besides job satisfaction it is worth looking at outcomes like satisfaction with life and satisfaction with family life. Figure 1 displays the results of this exercise. It compares satisfaction with job, life and family life from about 1 to 2 years before the job change to the first year afterwards. In addition, we distinguish between two types of job change, namely job changes after one’s own resignation and after plant closure.

Job satisfaction of people who switched jobs voluntarily soars up. The rise is tremendous: It roughly equals the effect of a trebling of one’s monthly net wage. At the same time, life satisfaction increases to a lesser extent and satisfaction with family life does not seem to change at all. In sharp contrast, workers who were forced to change employers show no increase in their well-being, but become much less satisfied with family life.

Figure 1. The well-being ‘effect’ of a recent job change


Note. Bars represent differences from individual mean values in satisfaction with job, life or family life in the first year after a job change. Each satisfaction measure is a self-assessment on an 11-point scale from 0 to 10. The differences originate from OLS regression estimations taking into account individual-fixed effects, socio-demographic and job characteristics. Whiskers denote 95% confidence intervals.

These results imply that firms should take the reason for the job change into account when they decide about measures of organisational socialisation. The successful onboarding of workers who did not choose to leave their previous employer may demand much more time and effort compared to those who intended to move. The intentions that ended the initial job are less clear with employees who mutually agreed to terminate the relationship with their initial employers or were dismissed for any individual reason. As our further analyses reveal, these types of job changes consistently yield job satisfaction increases lying between the extremes of quits and company closures. In addition, our findings show up independently of workers’ age, gender and education.

An interesting facet of the German labour market is that most people appear to switch jobs voluntarily and consequently experience great satisfaction with their new job. The most important question for firms that want to retain certain employees is probably how long the extraordinarily high job satisfaction of those happy newcomers lasts and at which point in time they might intend to change jobs again.

Previous research on managers describes a pattern called the ‘honeymoon-hangover effect’ of job changes. Accordingly, the high wellbeing at the beginning of a new job does not seem to last. To check whether this phenomenon concerns the working population in general, we exploit our representative dataset to follow workers over time before and in particular after their job changes. Figure 2 shows that a decrease of job satisfaction precedes quits, which aligns with the notion that a high level of work-related wellbeing remedies turnover at the firm level. Afterwards, job and life satisfaction increase but decline sharply afterwards. Within one year, life satisfaction returns to its initial pre-job change level while even the extensive honeymoon effect in job satisfaction is largely gone. According to additional analyses, job satisfaction continues to decrease during further years of job tenure, though at a reduced rate.

As tremendous satisfaction rises after changing employers, the “magic of the new“, to quote poet Hermann Hesse, disappears just as quickly. The crucial challenge that firms face is therefore how to maintain the euphoria of the fresh start once the organisational socialisation has been completed.

Figure 2. Well-being before and after job changes triggered by quits


Note. The orange (blue) line displays differences from individual mean values in life (job) satisfaction around a job change. The differences originate from OLS estimations taking into account individual-fixed effects, socio-demographic and job characteristics. Both satisfaction measures are self-assessments on 11-point scales from 0 to 10. Job changes take place from t = -1 to t = 0. The points in time t denote SOEP interviews. The period of time between two interviews is approximately one year.



IMG_4541Adrian Chadi ( is research associate (Post-Doc) at the Institute for Labour Law and Industrial Relations in the European Union (IAAEU) and the University of Trier. He received his PhD from the University of Münster. His research interests include behavioural economics, experimental economics, labour economics and personnel economics. Among other outlets, his work was published in Management Science and European Journal of Political Economy.

clemens-hetschkoClemens Hetschko ( is research associate (Post-Doc) at the Department of Economics at Free University Berlin, where he also obtained his PhD. His research mostly covers applications of happiness research to public economics, labour economics and personnel economics. He published, for instance, in The Economic Journal and Oxford Economic Papers.