LSE - Small Logo
LSE - Small Logo

Sima Sajjadiani

John Kammeyer-Mueller

Alan Benson

March 13th, 2024

When a high performer leaves, the firm loses more than a good worker

0 comments | 12 shares

Estimated reading time: 5 minutes

Sima Sajjadiani

John Kammeyer-Mueller

Alan Benson

March 13th, 2024

When a high performer leaves, the firm loses more than a good worker

0 comments | 12 shares

Estimated reading time: 5 minutes

Losing employees is tricky for organisations, especially when high performers are the ones leaving. Sima Sajjadiani, John Kammeyer-Mueller and Alan Benson studied possible domino effects of employee terminations. They investigated who is likely to leave following the departure of high performers and found that future turnover depends on the circumstances under which they left.


Losing high performers is costly at the outset and has spiralling consequences on the business over time. Finding and training new hires is time-consuming and costly, and there is a strong possibility that the new hire will still fall short of the performance of the departing employee.

In our recent study, we ask who is likely to leave following the departure of high performers, and whether future turnover depends on the circumstances under which high performers (and others) depart. Our interest in this question is motivated by the concern that remaining workers will be affected when a high-performing colleague departs and the circumstances that led up to it.

When a high performer leaves for other job opportunities, other high performers might reasonably speculate, “Maybe there’s another option out there for me.” When a high performer leaves because of dismissal or layoff, remaining high performers might reasonably ask, “Could I be next?”

To examine the consequences of losing a high performer, we analysed staffing and performance dynamics at 1,620 retail stores, an attractive setting because retail store locations are relatively insulated from each other, and performance and staffing are relatively easy to track. This means we could examine departures as part of the business’s dynamic system rather than as individual events. The number of observations across multiple time points allowed us to estimate an impulse response following departures (essentially the shockwave of turnover that ebbs after a few months).

We estimated these impulse responses for high and low performers and based on the circumstances surrounding their departures. We could thus evaluate how voluntary turnover (quits), dismissals, and layoffs at an earlier period differentially influenced the subsequent departure of remaining employees over many months.

Our main finding was that the departure of high performers, both voluntary and involuntary, causes a domino effect among the remaining high performers, resulting in increased quitting compared to other groups. For organisations seeking to build high-performing teams, perhaps the most alarming result is that voluntary turnover among high performers corresponds to a substantial increase in voluntary turnover for other high performers. For the three months after a high performer quits, turnover rates for other high performers rise six per cent per month, or a cumulative 18 per cent, before recessing to normal levels. Turnover for low performers, in contrast, declines, suggesting the business as a whole will shift to lower performance beyond the immediate effect of losing the high performer.

Turnover patterns at the businesses we studied also exhibited other general patterns. In the months following the initial departure event, voluntary and involuntary turnover tend to beget later voluntary turnover among the same types of workers. Layoffs are an exception: a layoff for anyone prompts greater departures for both high and low performers.

We also found that the dismissal of high performers is consequential for the human capital of the business both by increasing the voluntary turnover among high performers and decreasing voluntary turnover among low performers. Interestingly, we found that after the dismissal of a low-performing employee, high performers were more inclined to stay as their turnover decreased. At the same time, the voluntary departure of low-performing employees increased.

High-performing employees likely view the dismissal of a low-performing colleague as a positive indication of the organisation’s commitment to maintaining quality standards. Low-performing employees may interpret it as a warning, leading to an increasing voluntary turnover. This is beneficial for firms on two fronts: it maintains a higher quality of human capital and averts potential legal challenges often associated with dismissing people.

Other studies offer clues as to why turnover spreads among similar workers. One possibility is offered by social comparison theory, which suggests that high performers compare themselves to other high performers and seeing their peers leave plants the idea in their heads as well. Another related possibility is presented by signaling: such departures are viewed by incumbents as a sign that the time is ripe for high performers to find opportunities elsewhere. The dismissal of a high performer could be seen as a breach of the psychological contract that high performers’ jobs are secure or lead high performers to conclude that their performance is not rewarded. When a low performer is dismissed, other low performers may voluntarily depart in anticipation of possibly being dismissed themselves.

Performance is only one dimension over which turnover may propagate. People may be more likely to depart if they share a friendship or other commonalities with departing workers. Departing workers, particularly managers, may try to bring their best workers with them. For these reasons, managers should be cognisant of the broader consequences that a departure could have on the business.

Retail is a fast-paced environment with notoriously high turnover. When a retail worker departs, evidently for greener pastures, workers typically face few barriers to finding other local opportunities themselves. However, the issues facing retail are fundamentally the same facing workers in a wide variety of industries.

When a high-performer leaves, that could lead incumbents to wonder whether their current employer is undervaluing their contributions relative to the broader labour market. It could also shift the social and operational dynamics of the business. As long as incumbents must adapt to a new normal, they may as well decide to do that in a new environment.

When high performers leave, they may explicitly solicit their former colleagues to join them. Although workers and managers have long brought teams with them, evolving anti-noncompete sentiment in the US and Europe will likely only grease the wheels of any workers thinking about making a move. In this environment, managers should be aware that the cost of losing a high performer isn’t just the cost of replacing that person in that job, but the ramifications could extend to the business.

Three main takeaways:

  • If someone resigns, gets fired or laid off, people from their work group are more likely to depart in the future.
  • Departure of high performers causes a domino effect among high performers, resulting in them quitting.
  • Layoffs cause the biggest post-event turnover, followed by co-workers quitting.

 


 

About the author

Sima Sajjadiani

Sima Sajjadiani is an Assistant Professor in the Organisational Behaviour and Human Resources Division of the University of British Columbia’s Sauder School of Business.

John Kammeyer-Mueller

John Kammeyer-Mueller is the Curtis L. Carlson Professor of Industrial Relations in the Department of Work and Organisations at the University of Minnesota’s Carlson School of Management.

Alan Benson

Alan Benson is an Associate Professor in the Department of Work and Organisations at the University of Minnesota’s Carlson School of Management.

Posted In: Labour | Management

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.