How Much Does a Good Boss Really Matter? (Ep. 107)

(Photo: Ben Dalton, Creative Commons )

Our latest Freakonomics Radio on Marketplace podcast is called “How Much Does a Good Boss Really Matter?”  (You can download/subscribe at iTunes, get the RSS feed, listen via the media player above, or read the transcript here.)

It’s based on a recent working paper called “The Value of Bosses” (abstractPDF) by Edward LazearKathryn Shaw, and Christopher Stanton. In the podcast, you’ll hear Lazear describe the basic problem:

LAZEAR: Suppose you look at a firm and you see that the firm is highly productive. Well, it may be highly productive because it has productive workers, because it has productive technology, or because it has good supervisors that are enhancing the productivity of the workers, and it’s not so easy to tease out one effect from another.

So how can you measure the impact of the bosses? Data, people, data. And Shaw came up with a huge data set from a company that included roughly 23,000 employees and 2,000 bosses. From their paper:

Three findings stand out. First, the choice of boss matters. There is substantial variation in boss quality as measured by the effect on worker productivity. Replacing a boss who is in the lower 10% of boss quality with one who is in the upper 10% of boss quality increases a team’s total output by about the same amount as would adding one worker to a nine member team. Using a normalization, this implies that the average boss is about 1.75 times as productive as the average worker. Second, boss’s primary activity is teaching skills that persist. Third, efficient assignment allocates the better bosses to the better workers because good bosses increase the productivity of high quality workers by more than that of low quality workers.

The podcast also includes a conversation we had a while back with a very different sort of boss: Joe Maddon, manager of the Tampa Bay Rays. He showed up briefly in our “How Much Does the President Really Matter?” podcast but here we have time for a more substantial conversation.

Finally, here’s an interesting related e-mail from a reader named Max Lotstein:

A friend of mine works for a company with an interesting way of rewarding its employees. Immediately after he was hired after graduating in May, my friend, Jon, was told, “Get ready for an awesome Christmas party!” And from what he told me, the party didn’t disappoint: it was phenomenal. After he related this story to me, I speculated that the over-the-top party, and its emphasis, were intentional, and part of a managerial strategy. Though the company could have redistributed the money for the lavish party (employees were flown in from all over the world, at great cost, among other things) and padded everyone’s salaries, that type of reward often goes unnoticed. Perhaps it’s like what Al Gore said in An Inconvenient Truth — that we are all frogs in slowly boiling pots of water, unable to recognize incremental changes [Ed.: see here]– and that by spending this money all at once, it creates a salient thing to recall whenever one considers the question “Does my company value me?”


Only an economist would consider the conclusion that a good boss matters for productivity an achievement. This is something anyone who has actually worked would realize. These results are based on one data set that is not generalizable so the magnitudes tell us nothing of the actuall effect. Bravo for achieving nothing. Disappointed with this episode.


the avg boss is 1.75 X as productive as the avg worker- wonder how much more does the avg boss make in wages compared to the avg worker


Generally I would think less than 1.75x the wage of those who work under them.

Only at the top tier of incomes does the difference jump.


Perhaps I missed something, but it appears that this research has failed to consider the null hypothesis. If the productivity of a work team with a bad boss is 1.0, and that of one with a good boss is 1.75, what's the productivity of a team without a boss? 0.75, 2.0, or somewhere in between?

Eric M. Jones

Ask yourself how much a really sucky boss matters. Well then, there ya' go.......


In my experience, a "really sucky" boss can have a considerable negative effect on productivity. If I had to guess, if my productivity working alone on some task was 1.0, then a good boss might raise it to 1.25, while a really bad one would reduce it to 0.5.

Leigh Branham

Let's not forget about the influence senior leaders have on direct managers. They create the culture and set the example about how to treat people that "bosses" follow. My own research on employee engagement based on analysis of 2.1 million engagement surveys from 10,000 employers indicates that trust and confidence in senior leaders is now a stronger driver of engagement than the direct manager. Towers Watson and Kenexa have come to the same conclusion.


People work only for their immediate bosses. Everything else is secondary from a productivity standpoint.

In any case, in many workplaces, the methodology used to measure productivity is very questionable.


If the worker doesn't mind being a sheep while deflecting responsibility to the boss, then a "bad" boss for some might be good for others. If you hire a "sheep", but get a "wolf"......well, good luck. If some joker thinks they are a better human for commanding others, this self-absorbing personality will come forward soon. And when a boss says "I work soo hard and am soo under-paid" when you both know they make 3x your salary.....good luck with that one.

The `Boss'

Dear Johnny;

You seem so angry. About what I am not sure? As far as I am concerned, the only one whose fate I can control is my own and I am working at that. I have learned the hard way that money does not make a person happy. The happiest moments of my life had nothing to do with money. Indeed, a few had to do with work. When I succeeded in teaching a class my lesson for the day, week, semester my way...I felt great. Like Wow, I can teach in a way that my students understand. So, the fact that I am not teaching now... and won't be until I finish off a writing project... is very hard for this teacher and scientist. Just something i have to live with for now. people I know want to know when I will be done? They just don't get what is required. Why should they? My real friends don't ask! I am reminded here of what an old friend once said... sticks and stones ... but names will never hurt me. And they won't.


Meg Greer

I just listened to your podcast about bosses, and would like to offer these comments from my experience as a boss and an employee.
As a boss, the two most important functions are to align incentives for your employees with the goals of the enterprise, and to provide the policies, systems and tools needed for employees to do the job.
When I ran a small business (Gymboree franchises to be precise), the most important factor in our success was enrollment in our toddler activity classes. Most franchisees paid their teachers an hourly or per class wage. I paid my teachers a wage, but I added incentive pay based on class enrollment. This meant that better teachers, who attracted, engaged and re-enrolled more toddlers (and parents) got paid more. I also gave them the tools they needed to keep up enrollment – an updated registration list with expiration dates for each member, and a credit card machine to collect fees. Simple and effective.
Now I am an employee in a large corporation in a highly regulated industry. I want my boss to advocate for me and smooth out bumps in the road to my productivity. A good boss escalates and resolves problems that stand in my way. He or she understands my job, communicates up and down the chain of command on my behalf, and works with management to shape policies and systems that improve my productivity.
Rather than holding up the regulations as a barrier, for example, I need management to provide policies and systems that help me comply with the laws without slowing down my progress. A lot of money goes into computer systems, but frequently they are designed without a sense of work flow. An example is that we have limited email storage, but need to keep extensive notes on communication with clients. However, our notes function does not interface with email, so we must cut and paste our often lengthy email exchanges into the notes. This requires expensive talent to perform mindless work that the computer could do.
That’s enough for this morning. I’ve got to go to work. If I were looking for the differences that make a good boss, these are the areas that I would examine.



Second, boss’s primary activity is teaching skills that persist.
can anyone explain this further? boss teach, boss engage inhouse trainer or external trainer. or merely boss encourage learning?

Ben S

I think Freakonomics Radio presents an intriguing question which has relevance to almost all forms of work where merit is based on some form of workplace hierarchy. However, I wonder about how valid a study can be when based on a fairly subjective set of words. This runs right down the lines of my “good” boss might be another’s “fair” boss. Just as one’s “fair” boss might be another’s “poor” boss.
In my experiences I feel I’m with many who have been fortunate enough to have great bosses as well as pained though dreadful bosses. In both experiences I would agree with Lazear in that a “good” boss leads to more productivity; however I just wish there was a better way to test it. The problem might stem all the way down to the idea that if one doesn’t like their job, they won’t enjoy their work, and even a “great” boss might come off as a “poor” one though the eyes of the disgruntled employee.


Dr. Will Felps

I think this is a great piece of research. Because the company somewhat randomly rotates bosses and employees (which is very unusual) is allows for some really strong inferences about causality. I am not aware of any other study that does this.

However, the conclusions that I would draw from the empirical results would be different from those drawn in the piece media and by the article.

The main question is: Does having a good boss make a big difference?
The Answer: According to the abstract for the article:(
"There is substantial variation in boss quality as measured by the effect on worker productivity. Replacing a boss who is in the lower 10% of boss quality with one who is in the upper 10% of boss quality increases a team’s total output by about the same amount as would
adding one worker to a nine member team."

This means that a boss in the top half (how they define "Good Boss") gets their workers to perform 11% better than a boss in the bottom half. That's important, but less than you might expect given our focus on leadership. In comparison, an employee in the top half of the IQ
distribution usually performs about 50% better than one in the bottom half (if memory serves).

Now, the company in the sample is one that handles very routine, phone-based service. I could imagine that the effects of having a good boss would be higher in less standardized environments... where the technical infrastructure and bureaucracy serve as substitutes for

The authors also show a "residual effect" of having a good boss on future performance (i.e. after the boss has left the workers continued to perform better for a while. The authors interpret this as reflecting that the key job of a boss is acting as a good teacher.

Perhaps... but given that most of the residual effect wears off relatively quickly (an indeed, only 14% of the "good boss" effect lasts more than a year: see footnote two in the above url, pg. 28)... my interpretation would again be different. Specifically, the results might be better explained as indicating that a good leader instills productive work habits and high expectations in employees. These are likely to last a while after the boss leaves, but not that long. In other words, I think any learning/teaching effects should have persisted... whereas good work habits would deteriorate more rapidly.

Anyways, thanks for the great work!



Given the experimental procedure described, i would have to make the same observation i make about pretty much all non-double blind studies, which is that there are confounding factors, ascertainable and unascertainable, that could just as easily account for all of the results. For instance, some managers might be in a position to get onto the more effective projects (through nepotism, their own insight, or some other means), which make their employees appear more effective either in terms of return on investment, or by allowing leverage of different technologies.


One angle I'd like to share, because I think is of utmost importance and defines a good boss, it is the "team productivity" rather than the individual productivity. a good boss brings the team together and assigns the right tasks to the right people, make sure that 1+1 =3, 2 to say the least but not less! I find it interesting that the article says good managers should be assigned to good workers, where I found from my own managerial experience that actually high achievers need space and they don't like the boss breathing under their neck. I even let them deal straight with upper management and only step in if necessary. they still need the guidance of a good boss but much less than average or poor performers. I'd say if the manager succeeds in showcasing the great people as role models for "good" or average workers, and either manages up the bad ones or manages them out, the team will do wonders. I actually find motivating and engaging the average people to be the hardest, they are not too bad to be micro-managed and not good enough to be decently rewarded, especially in large corporates during the craze of slashing costs these days



What music is that at 6:00 minutes??? I need that in my life.

Bradley Calder

Please explain the Valve corporation.

Thad Humphries

"A small matter," said the Ghost, "to make these silly folks so full of gratitude."

"Small!" echoed Scrooge.

The Spirit signed to him to listen to the two apprentices, who were pouring out their hearts in praise of Fezziwig: and when he had done so, said,

"Why! Is it not? He has spent but a few pounds of your mortal money: three or four perhaps. Is that so much that he deserves this praise?"

"It isn't that," said Scrooge, heated by the remark, and speaking unconsciously like his former, not his latter, self. "It isn't that, Spirit. He has the power to render us happy or unhappy; to make our service light or burdensome; a pleasure or a toil. Say that his power lies in words and looks; in things so slight and insignificant that it is impossible to add and count 'em up: what then? The happiness he gives, is quite as great as if it cost a fortune."

--Charles Dickens, "A Christmas Carol", Stave II