Most organisations say they promote on merit. But what if the very way we define merit is quietly undermining leadership quality?
In 2020, Alan Benson, Danielle Li and Kelly Shue published a large scale empirical study in The Quarterly Journal of Economics examining promotion decisions across 131 firms and almost 40,000 sales employees. Using detailed performance and organisational data, they tested a long standing idea in organisational research known as the Peter Principle, the tendency for organisations to promote people based on success in their current role rather than suitability for the next one.
Their findings offer a sobering challenge to common promotion practices.
The researchers found that high performing salespeople were significantly more likely to be promoted into management roles. On the surface, this appears fair and rational. Strong performance is rewarded, effort is incentivised and decisions feel objective. However, when the authors examined what happened after promotion, a different pattern emerged.
Sales performance before promotion was negatively related to managerial effectiveness after promotion. In other words, those who excelled most strongly as individual contributors tended, on average, to be less effective managers. Managerial effectiveness was measured by the extent to which managers improved the performance of their teams over time, controlling carefully for firm conditions, team composition and individual differences.
Crucially, this was not simply a statistical curiosity. When the authors compared workers who were promoted on the margin, those promoted mainly because promotion rates were unusually high at that time, they found that marginally promoted lower sales performers were better managers than marginally promoted top sales performers. This suggests that organisations were systematically lowering the bar for managerial suitability when candidates had strong individual performance records.
Why might this happen?
The authors argue that promotion systems often serve two competing purposes. One is to place people where they will perform best. The other is to motivate effort by signalling that high performance will be rewarded. Promoting top performers sends a powerful message about fairness and effort, even if it comes at the expense of leadership quality.
From an inclusion perspective, this tension matters. When promotion criteria prioritise visible, individual outputs, they can inadvertently favour roles, behaviours and individuals whose contributions are easiest to measure. This risks overlooking people whose strengths lie in collaboration, development of others and system level thinking, capabilities that are essential for inclusive leadership but often less visible in performance metrics.
The study provides an important example. The authors found that collaboration experience, measured by how often individuals worked with others on shared sales outcomes, was a positive predictor of managerial effectiveness. Yet collaboration was not consistently rewarded in promotion decisions. In fact, individuals who worked more collaboratively were often less likely to be promoted, despite later performing better as managers.
This has clear implications for how organisations define and assess potential. If leadership roles require different capabilities from individual contributor roles, then promotion decisions based primarily on current performance risk reinforcing a narrow model of success. Over time, this can shape leadership populations in ways that limit diversity of style, perspective, and approach.
The authors estimate that if firms promoted the best potential managers rather than the best individual performers, team performance could improve by around 30 percent. They are careful not to argue that organisations are irrational. Instead, they highlight the real trade offs involved and the need to make them explicit.
For organisations committed to inclusion, the message is not to stop rewarding performance. It is to think more carefully about what is being rewarded, how potential is assessed and whether current systems truly support the kind of leadership the organisation needs for the future.
You can read the original article here: https://academic.oup.com/qje/article/134/4/2085/5550760
