Media Briefings

SOCIAL INTERACTIONS IN THE WORKPLACE: The impact of co-workers on job satisfaction in the UK

  • Published Date: April 2014

Individuals’ job satisfaction goes up when there is a larger fraction of male and older workers in the workplace and when there is a smaller fraction of high earners against whom to compare pay. These are among the findings of research on the impact of social interactions in the workplace by Semih Tumen and Tugba Zeydanli, to be presented at the Royal Economic Society’s 2014 conference.


The authors analyse data from the Workplace Employment Relations Survey and the British Household Panel Survey to examine the impact of aggregate job satisfaction in the workplace and the local labour market on individual-level job satisfaction. They find that an increase in aggregate job satisfaction level leads to three times higher increases in individual-level job satisfaction at the workplace than in the local labour market.


‘My job satisfaction level increases with the job satisfaction levels of my co-workers and also the workers with whom I share similar local labour market conditions.’


In most research, job satisfaction is documented as being positively correlated with worker performance and productivity. But whether there are any visible footprints of social interactions in job satisfaction or not is an unanswered question.


How do my colleagues’ job satisfaction levels in the workplace affect mine? To what extent does the job satisfaction of people with whom I am working within the same industry and region influence my job satisfaction? What kind of group-level interactions can affect my job satisfaction level? Answers to these questions may be important for designing policies to increase job satisfaction.


This study examines whether aggregate job satisfaction level in a certain labour market environment can have an impact on individual-level job satisfaction. If the answer is yes, then policies targeted to increase job satisfaction can increase productivity not only directly, but through social interactions too.


The researchers conduct their analysis at two aggregation levels using two different datasets from the UK. First, they use the 2004 Workplace Employment Relations Survey (WERS) to test the existence of job satisfaction spillovers at the workplace level. In the workplace-level analysis, the reference group that the social forces are effective is the set of workers in each workplace.


Second, they use the 2004 British Household Panel Survey (BHPS) to form industry-region cells for the purpose of testing the existence of spillovers at the local labour market level. In this second exercise, they try to capture social processes that involve collective aspects of community and work life. They conclude that there are sizable social interactions in job satisfaction that should not be ignored in assessing policy effectiveness.


An increase in aggregate job satisfaction level leads to three times higher increase in individual-level job satisfaction at the workplace than at the local labour market. Contextual social effects also have a significant impact on individual job satisfaction level.


At the workplace level, job satisfaction at the individual-level goes up when a larger fraction of male and older workers are present. At the local labour market level, the findings say that individual-level job satisfaction score goes down as the fraction of workers with greater access to promotion opportunities goes up in each industry-region cell.


The researchers also document that there are significant ‘income-comparison effects’ at the workplace, but not at the local labour market. In particular, they find that individual-level job satisfaction goes down with average earnings and the fraction of high earners – that is, those who earn above the median wage within the relevant worker population – in the workplace.


These results suggest first, that there are large gains to policy interventions to increase individual-level job satisfaction, as there are significant positive feedback effects from group-level job satisfaction toward individual-level job satisfaction in the form of spillover externalities.


Second, they suggest that failing to account for the spillover externalities in job satisfaction may lead to an incorrect assessment of the effectiveness of job satisfaction policies. Thus, policy-makers should internalise these externalities.


Third, job satisfaction spillovers are much stronger at the workplace level than local labour market level: therefore, designing and enforcing job satisfaction policies at the workplace level will likely be more effective than implementing such policies at the local labour market level.





Notes for editors:

‘Do social interactions affect job satisfaction?’ by Semih Tumen and Tugba Zeydanli


For further information, contact:

Semih Tumen,, +90 532 343 5920

Romesh Vaitilingam:, +44 7768 661095