Skilled managers boost employee retention, but not necessarily productivity, study finds

Credit: Hipanolistic for iStock

Supervisors with good people-management skills keep employees happier and less likely to quit, but might not spur higher productivity, a new study has found.

In a paper forthcoming in The Journal of Political Economy based on research at a large tech company, Berkeley Haas Prof. Steve Tadelis and co-author Mitchell Hoffman of the University of Toronto found that employees who were assigned to managers who scored high on people-management skills were less likely to leave the company than those assigned to less-skilled managers.

Prof. Steve Tadelis
Prof. Steve Tadelis

“The success of companies depends crucially on the relationships between managers and their subordinates, but whether or not managers specifically influence employee performance or decision-making has not been well-studied,” Tadelis said. “We found a very clear connection in terms of a company’s ability to retain its best employees.”

Tadelis and Hoffman gathered data from surveys of employees working in a variety of jobs, such as engineering, finance, or marketing. The company conducted annual surveys that asked the employees about a range of topics—from whether their manager was trustworthy to whether the manager provided adequate coaching—and used the responses to create a measure of the supervisors’ skills. Along with personnel data on the employees, the researchers then analyzed the extent to which good people-management skills affected several employee outcomes.

Turnover is a critical issue in the tech sector, since replacing workers is difficult due to the specialized knowledge many jobs require. “There’s tremendous competition over talent in tech,” Tadelis says. “Losing good people is really painful.”

There’s tremendous competition over talent in tech. Losing good people is really painful.

Their results showed that, indeed, supervisors with good people-management skills kept employees from quitting their jobs. Specifically, if an employee moved from a manager ranked in the bottom 10% of the people-management score to a manager ranked at 90% or above, they would be 60% less likely to leave the company, the researchers concluded.

They also looked at resignations from highly valued employees to determine whether the firm “regretted” the loss. Firms keep information on which employees are top performers in order to determine future promotions or for rehiring, if the company tries to incentivize someone to return. The researchers found an even stronger association between managers’ scores and regretted quits, showing that good managers prevented good employees from leaving the company.

The results were also stronger for employees at higher levels of the company, suggesting that a manager’s people-management skills are particularly important for employees who may have more demanding jobs.

Interestingly, they found that supervisors’ people-management skills were not associated with other measurable outcomes. The authors found no evidence that better managers helped spur more employee patents, or led to better subjective job performance or a higher probability of promotion. The authors interpret these results as evidence that better people management skills may not make workers more productive, but do improve their job satisfaction. Tadelis cautioned that at a high-tech company, it is difficult to measure individual productivity, so there may be other benefits to having better people-management skills that could not be measured in the study.

The reserachers also found that managers with good people skills were viewed more favorably by the company and rewarded for their abilities. A manager at the 90th percentile of the people-management score was three times more likely to be promoted relative to a manager at the 10th percentile. This finding, along with the result that good managers were more impactful at higher levels of the company, indicates that the company was effective in putting good managers into positions where they could make the most impact.

Finally, the authors calculated that if a manager moved from the 10th to the 90th percentile of the management score, employee costs would be reduced by 5% due to the lower rates of turnover, and the company would benefit dramatically from these savings

“More and more companies are introducing more sophisticated tools for human resource analytics,” said Tadelis. “Our research shows that HR data and surveys that are gathered by some companies can be used effectively to improve HR outcomes.”