Managers are everywhere, yet what they add to an organization is surprisingly hard to pin down. Recent economics research offers a clearer picture, and it points to a few practical priorities for anyone who leads a team.
This article draws on two working papers from the National Bureau of Economic Research (NBER), a US research network, and turns their findings into questions a leader can ask about their own week.
Why is it hard to say what a manager adds?
A manager rarely produces the final product. A salesperson closes the deal, an engineer ships the code, a nurse treats the patient. The manager’s contribution shows up indirectly, through the output of other people, which makes it difficult to measure and easy to underrate or overrate.
Economists who study organizations have tried to close this gap by combining theory with real personnel data. A December 2022 NBER working paper by John Roberts and Kathryn Shaw reviews this literature, with a focus on the middle managers who sit between top executives and front-line employees. A January 2025 paper by Alan Benson and Kathryn Shaw, titled “What Do Managers Do? An Economist’s Perspective,” builds on that work and tries to organize it into something usable.
What does the research say managers do?
According to the Benson and Shaw paper, managers coordinate individuals toward shared goals. The authors describe several ways managers add value:
- Hiring, choosing who joins the team.
- Retaining, keeping capable people from leaving.
- Training, helping people build skills.
- Monitoring and evaluating, watching progress and judging results.
- Allocating, deciding who works on what.
- Supervising, day-to-day oversight of work in progress.
None of these is surprising on its own. The value of the framework is that it separates managing from simply being senior. A leader who spends the week in status meetings may be doing a little monitoring and no training, hiring or retention work at all.
People managers or project managers?
The paper also draws a distinction between “managers of people” and “managers of projects.” The first group is responsible for the growth, retention and performance of individuals. The second is responsible for getting a particular piece of work delivered, often across people who report elsewhere.
Many real jobs blend the two, but the split is a useful diagnostic. If a manager is judged only on project deadlines, people development tends to lose out when time is short. If a manager is judged only on team morale, deadlines can slip. Being clear about which hat the role wears most of the time helps set expectations for both the manager and the team.
Do the right skills depend on the kind of work?
The authors suggest that managers should weigh these skills differently depending on the production technology they work within, meaning the way work actually gets done. Think of how different an assembly line, a software team and a consulting practice are.
- Where work is repetitive and measurable, monitoring and allocation may matter most.
- Where work is creative or knowledge-heavy, hiring, training and retention often carry more weight, because quality depends on individual judgment that is hard to inspect from outside.
This is a framing, not a rule. But it explains why advice that works in one setting, such as tight tracking of daily output, can backfire in another. A leader who has moved from one kind of team to another may need to change which of the six activities gets the most attention.
Do people management skills reduce turnover?
A separate NBER paper tackles one of those activities directly. Mitchell Hoffman and Steven Tadelis, in “People Management Skills, Employee Attrition, and Manager Rewards: An Empirical Analysis” (issued February 2018 and revised April 2020), used personnel data from a large high-tech firm. They found that survey-measured people management skills have a strong negative relation to employee turnover. In plain terms, teams whose managers scored better on people skills tended to lose fewer employees.
Two details from the same paper are worth noting:
- The authors report that these skills do not consistently improve most of the other observed outcomes besides attrition. Good people skills are not a magic lever for every metric.
- Managers with stronger people skills also received higher subjective performance ratings, were promoted more often and saw larger salary increases. In that firm, at least, the organization rewarded the skill.
The study covers one firm, so it does not prove the same pattern holds everywhere. Still, it supports a practical point explored in leading a team when few people quit: retention is not a side effect of good management, it is part of the job.
Where should a manager’s time go?
Turning the framework into a weekly habit is simple enough to try. A worked example, using illustrative numbers rather than research data, shows how.
Suppose a manager has a 40-hour week and tracks it for two weeks:
| Activity | Hours (typical week) | Share of 40 hours |
|---|---|---|
| Monitoring and status updates | 14 | 35% |
| Allocating and planning work | 8 | 20% |
| Supervising and unblocking | 8 | 20% |
| Training and one-on-one coaching | 4 | 10% |
| Hiring and onboarding | 2 | 5% |
| Retention conversations and recognition | 2 | 5% |
| Other administration | 2 | 5% |
The hours add up to 40, and the shares add up to 100%. The pattern is common: monitoring crowds out the developmental activities. Whether that is the right balance depends on the work. A manager of a repetitive operation may be comfortable with it. A manager of a design team who spends only 10% of the week coaching, in a business where skills drive quality, has a clear signal to rebalance.
How can a leader shift the balance?
Small changes tend to work better than a big reorganization.
- Replace some monitoring with trust and clear goals. Shared dashboards or short written updates can replace status meetings. Meeting design is covered in how to run meetings that respect time.
- Make feedback routine. Coaching is easiest when it happens in small, frequent doses, as described in how to give feedback people can use.
- Protect time for the people questions. Block recurring one-on-ones and treat them as non-negotiable, the way a deadline would be.
- Decide what you will stop doing. Every added hour of coaching has to come from somewhere, and allocation decisions sometimes can be delegated to the team itself.
What should a leader measure about their own management?
Because a manager’s output is indirect, it helps to track a few indicators of the activities above. Examples include how long open roles take to fill, how many people leave the team each year, whether team members can name a skill they have built in the last six months, and how clearly people understand priorities. None of these is perfect, and the Hoffman and Tadelis findings are a reminder that a single outcome rarely tells the whole story. The aim is to notice drift, not to score yourself.
Distributed teams deserve special attention here, because monitoring by observation is harder and trust has to be built on purpose, a theme in building trust in distributed teams.
A better question for your next planning week
The research does not hand leaders a formula. What it offers is a vocabulary: hiring, retaining, training, monitoring, evaluating, allocating and supervising, applied differently depending on how the work gets done and whether the role leans toward people or projects. The practical step is to look honestly at where your hours go, compare that with what your team’s work really demands, and adjust one thing at a time. The evidence that people skills are linked to lower turnover is one more reason to make sure that part of the job does not get squeezed out.
This article is general information about management practice, not professional, legal or business advice. Research findings describe the specific settings studied and may not apply to every organization.
Sources: National Bureau of Economic Research
