Ted Audio Collective.
Hi, this is me here at ASI and you're listening to HBS After Hours.
I'm delighted tonight to be talking to my colleague, Raffaella Sadoun, who has done spectacular work.
And we're going to be talking a little bit about her work on the value of management.
So welcome, Raffaella.
Thank you. It's a pleasure being here.
Just as a means of introduction, you know, you've created this enormous body of work with Nick Bloom and John van Rienen and others.
And a lot of it is summarized in this fantastic HBR piece that you guys did last year, which I think just won the HBR McKinsey award for best article in the HBR.
And it's called, why do we undervalue competent management?
And it really kind of is a business school Dean's dream come true.
You know, which is what we do and everything that we teach matters.
But you kind of frame it in an interesting way, which is you think we've come to devalue management and this whole body work suggests that we're wrong about that.
So tell us a little bit about why you think we came to devalue it and what we get wrong in that process.
So I think the title and the spirit of the article are actually very much inspired by my experiences teaching here.
So when I got here, this was already eight years ago.
Um, I started teaching the basic RC strategy course, which is the first year course, uh, that every MBA students go through.
They cannot choose.
And one of the key principles of what we teach is that, um, what it's called operational effectiveness, which is really the execution piece, you know, the processes, the routines are certainly very important for success, but they are easy to imitate and therefore their immittability means that they
cannot be a source of sustainable competitive advantage and therefore strategy matters.
And therefore, you know, you should as an executive, put more emphasis on your work as a strategic decision maker, which means understanding the business in which you want to play and how you position yourself within that business.
So that perspective to me, um, really sort of shut down all the potential complications that come once an executive has to execute, right?
I've often heard of this referred to as it's table stakes.
Operational effectiveness is almost a given, then we talk about the interesting stuff and you're suggesting that that's quite incomplete.
Well, you know, when I came here, I had just started looking at the data, which showed that it was tremendous variation in adoption of basic management practices.
And that variation was substantial, not just across countries.
As you might expect, to some extent, management is an investment that you make.
You know, it's easier perhaps to find managerial resources in developed countries relative to developing countries.
But the surprising thing is that there was tremendous variation within countries as well.
When you say tremendous variation, Rafaela, you had to first actually define what management was.
That's like the first step, right?
So how did you go about that?
Or how did you think about that?
This was something, this is a project that is called the word management survey, which you know, is the data that is at the, you know, the foundation for this HPRP study described, is a project that started in, uh, 2004 very much inspired from the private sector.
Uh, we started, uh, working with, uh, consultants in McKinsey.
They believe that they had understood basic processes that they thought based on their experiences was systematically associated with, uh, firm level growth and firm level success.
They had not validated these ideas, however, in large samples.
And I was part of a group of economists that, uh, decided to say, okay, you know what, if you really believe that that's true, let's see, right.
Let's measure it at scale.
Right. So the definition of what good looks like came from the private sector, came from a, from a survey greed that was already defined 18 processes that they thought were important.
And I was part of a group of academics that said, okay, let's scale that measurement up and to this date, we have 15 ,000 organization in, in, uh, 34 countries that have been measured using this survey tool.
That's amazing. And it's not just measured, right?
I mean, there's actually been interviews, qualitative interviews, right?
So it's not just kind of some ticking the box, right?
No, absolutely not.
I mean, there is a whole world behind the measurement, which actually I think I loved it, I still love it because it's, you know, you think about academics as being holed up in their office.
Here it was, I was actually managing teams myself who were in a community.
We were all based at the London school of economics at that point.
This was during my PhD years.
We took a conference room of the, of the school where typically you would have discussions of equations and stuff.
We converted that room in a call center.
Very good. And it was, you know, for two months we, we worked with the MBA.
Typically it was MBA students in between classes, and it would happen, this would happen in the summer and we would choose them.
We would motivate them.
We would incentivize them, but we would work together.
And train them to run interviews that were fundamentally open -ended talking with plant managers of manufacturing companies that were randomly selected.
That was something that we really wanted to get.
You know, here at HBS, for example, you always hear about the winners, the big companies, but the economy is made of, you know, a whole variety of businesses.
So we decided to do, you know, representative sample of companies.
And so for the summer, they would cold, cold these companies, convince the plant managers to be interviewed for about 45 minutes to an hour and run this survey grid, which effectively was scoring the answers of the managers against a scale that we had predefined.
So you're really trying to quantify what it means to be well -managed.
Exactly. On a scale that is quite simple because, you know, one would be that there is very little adoption of that process.
And five means that there is a full adoption.
And these things are quite basic.
I call them housekeeping stuff.
Give us some examples of what those things are.
So for example, you know, one of the questions would have to do with how do you monitor production, right?
Again, it would be asked in an open -ended way, give you an example, you know, how exactly how you just did, and companies that were closer to one in our grid would have very rudimentary ways of monitoring what was actually happening on the shop floor or that data would not be visible to people so that it
wasn't actionable. On a five, you know, an example is a plant manager I spoke with actually who had access to the production data continuously, who could look at this data from his bedroom.
I mean, I don't know how happy his wife was about that.
But you know, it's this idea of really believing in the data and making sure everybody had access to it, could see it, would be visible.
And you know, it rate editing questions range from stuff like that that is very operational to things that, for example, have to do with human resources.
For example, how do you promote people?
It's one of those questions in certain places.
Promotions are actually, you know, not really related to how good you are.
Sure, sure. So you gather all this data on, I think you said 15 ,000 firms, 30 plus countries.
And what are the first kinds of things you observe in terms of just the variation and then what drives the variation?
Well, the first thing you observe is the variation, which is I know it must sound surprising, but think about it, that variation in principle, if things are immutable should not exist.
Right. You know, and I think economists didn't believe that that would exist because, you know, for an economist, if you're badly run, you should be selected out of the market.
So, you know, the first thing is, wow, look, these companies can coexist.
The second piece, which I think is fascinating is how well these metrics, if you look at the data across countries, how well do they map with the metrics such as, you know, GDP per capita, which it's nerdy, I know, but it's very inspiring.
And the reason is fundamentally what this research agenda is suggesting that this basic management practice is a basic way of doing things affect firm level productivity with testing the data is strong correlations, but ultimately they affect country level productivity.
And so the reason why I say it's inspiring because I'm here, you know, teaching this stuff to people who could potentially have any impact that goes even beyond their business.
I mean, if you think about the biggest questions we think about, they're about why GDP per capita variations exist to the degree they do, why TFP or total factor productivity is so different across countries.
And, you know, that has always been like some disembodied notion of technology.
And it feels like you guys are digging deep into it and actually saying there's a huge chunk of that, which is actually management.
Yes, that's exactly right.
So now you've kind of uncover all this variation.
How does it map to performance exactly?
Like, how do you, how do we quantify that?
So the methodology was trying to be as rigorous as possible.
So we mapped this survey data with external metrics of performance that we found from, you know, there are lots of public accounts, especially in Europe.
This data is very rich.
So what we know in the data is that better management correlates with metrics of productivity, both labor productivity, total factor productivity, firm growth, firm survival, even interestingly, R &D spending and patents.
So it's a measure that seems to capture something nice happening within this firm.
These objective measures are then translating into real operational differences and large operational differences.
I think if you, I mean, as I recall, you go from the top decile to the bottom decile, you get large changes in profitability, large changes inside.
And profitability, of course.
Yeah, of course. The issue I'm having with some of this is that these are survey based instruments, right?
So if I think large firms are going to be higher productivity firms for all kinds of reasons, and we observe that in the data all the time.
And if we think that people in these large firms that are successful feel good about what they do, and then there are these small firms that are not that productive and they don't feel like good about what they do, because they're losing, is there a concern that the survey instrument is kind of picking
up these subjective ideas of we're kind of not doing that well or we're kind of really doing well so that when we grade ourselves, you know, we have a better sense of who we are?
Do you know what I'm getting at?
Yeah, no, I think, you know, that that is why it was very important to have full control on what the survey actually measured and how the questions were asked.
So I think you're hinting to several issues.
One is, do we get what's really happening on the ground or do we get their Exactly their impressions.
That's why we didn't tell in advance that we were actually scoring the questions.
We were not telling the managers we are evaluating you.
Interesting. Because we were concerned precisely that they would tell us, you know, what we wanted to hear or things that were actually not very related to what actually was happening.
And on the other hand, also the analysts didn't, the people who were running these interviews, they didn't know much about the company.
Right. So the survey was double blind to minimize biases of both ends.
And then who scored it?
Was it a different person yet who scored it or was the person who had the conversation.
Oh my God, they're opening, you know, I can't go on for hours.
No, no, no, but it's we implemented, I'm very proud of all these basic managerial processes, but you know, we actually did for every survey was monitored, double scored.
There was another person listening at the interview at the same time, so that we could have consistency and discussions.
And also we wanted to make sure that data was high quality.
That's what I love about this actually, because the survey based research is potentially quite rich, but you have all these issues and it feels like you guys thought about the real issues in the survey based design, which was really impressive.
And I think it's, you know, we learned as we went along, started small, we ended up scaling and we are at the point now in which the US census is actually using these questions in their own questionnaires with self -reported measures.
So I think that there is something to say about learning and improving.
So let's think about a little couple of the things that once you dig into this, you know, who does management better in some sense is an interesting part of this, right?
So first thing we know is there's a ton of variation, second at least to actual performance differences and then who's doing well.
And so there are a couple of really interesting things I was hoping to probe you on.
One is coming from a finance perspective, you know, we think about agency problems and in fact, what do we find?
Firms with wide public shareholder basis actually are the top performers.
Is that right? And in fact, private equity comes in behind them.
And then of course the worst players are the family players.
So did that surprise you or how did you...
So you know, to clarify the last ones are not so much the family owned are the family owned and the family managed?
Sorry, exactly right.
There are important distinction between ownership and control here.
Right. You can be family in the data.
What we show is that you, you know, companies that are owned by family, but run by professional managers actually do quite well.
The problem is in family owned and family CEO firms, and which contains also the founder CEO and founder owned and founder managed firms, which I agree with you from the perspective of the agency principle agent.
Why? Right? This guy should be super motivated and do the best for the firms.
And the disbursed shareholders should have the biggest problems.
So what's going on here?
I think two things are going on.
One is that often family ownership and family controls has a big trade off in terms of selection.
Yeah. So you, for example, you know, as you can hear from my accent, I'm not American.
I'm actually Italian where family CEO firms go strong.
Right. And, um, and the issue is the selection of who is a CEO is based often on who is the first son, for example, sometimes you get lucky, but sometimes you are restricting your pool of talent such that you're not actually capturing the most capable person.
Yeah. And so here is a problem of, uh, you know, who gets to run the company.
Perhaps sometimes you give priority to your family bonds relative to merit.
I think this is so interesting because there's a folklore about family owned firms, you know, that, you know, the first generation is okay.
And then the second generation it's like the wooden Brooks thing, but I hear you've really quantified that you've really actually made it evident in the data.
How do you make sense of the publicly disbursed shareholders doing so well?
I, you know, I really think that the, what we're capturing is investments that these companies have to make.
Right. It's you know, they are under pressures that perhaps companies, private companies are not.
Right. And also the fact that they can probably attract and compensate managers in a different way.
One of the best parts of this research, and there's a, you know, there's like 10 ,000 fascinating results.
I'm just going to pick out a couple of them.
Is, um, one of the titles of your articles is I think it's Americans do it better or do it better.
Well, but then you have another piece in HBR where, you know, you have three grudging European authors who confess that Americans are really good at management, right?
So tell us about that.
Tell us about how that shows up in the data and why you think that might be.
Um, that is a project that, you know, is it strange that I love my projects?
No, not at all. That was actually one of the first chapter of my PhD dissertation that what we found in that paper, what, what we argued in that paper, and we could actually quantify it and measure it, is that companies, multinational companies that were American, uh, in the data seem to get much more
in terms of productivity effects from their IT investments.
Now the setting here is that we were at a point in which there was a little bit of a mystery in the sense that we, you know, a macro economist knew that there was going on, a lot of, a lot of investments in ICT was going on, what's happening, however, the productivity returns of these investments had
not yet figured out, showed up in the day.
Okay. So once we measured and we, we saw that there was this strong difference between American companies and other companies in terms of returns from ICT investments, the big question is why?
And ICT is just information and communication technologies.
So this would be, you know, what we measured was quite basic.
Now it would be much more advanced, but it's, you know, hardware investments and software investments.
The thesis of that article is that, um, in order to see the productivity results of ICT, you need to have complimentary investments, uh, in place for those, you know, for this technologies to really deliver on your promises.
It's not enough to plug a computer in the wall, just to give you a concrete example, but you need to have a person that has the right skill, the right competences and can make use of the data.
This is, you know, well known in this literature that this complementarities exist, what we think is happening is that by having better management practices, you can create a better match between the technologies and the people who use these technologies.
American companies in our data had much more flexible human resource, uh, policies.
Yeah. They were more proactive in identifying who had the right skills and in moving people around such that the right skill could match the use of the technology.
And so as a consequence, you would see these strong differences in terms of productivity of information technology.
Right. How difficult was it for three European authors to admit that Americans do IT better?
Oh, personally, you know, I think, uh, not very hard, to be honest.
Very good. You got to follow the data.
I'm American now. So there you go.
There you go. There you go.
Um, so this is, I mean, spectacular.
And there's a whole bunch of research.
In fact, there's a website you guys now have, the world management survey .com.
Is that right? Or .org?
Both. And, uh, you know, we put everything there and it's a free disposal in the sense that we put the data as well anonymized, of course, and we would love for people to use the data, use the methodology and just, you know, go with it.
Yeah. So tell me where you're going now.
That was a huge body of work.
I think it was almost a decade's worth of work.
Spectacular. It has this kind of, the stuff we talked about is just in some sense, the tip of the iceberg.
Tell me what you're getting interested in now.
Well, two things. Uh, one is, um, very much again, influence from my experience here.
Uh, where, you know, there is a strong sense that leadership can have an effect.
Part of my research is trying to understand where do these differences in management come from and how can we relate it?
Can we relate these differences to the specific behavior of CEOs and differences in corporate culture?
So I have, um, um, you know, an agenda there that measures what CEOs do with their time at scale using similar methodologies and sort of maps these differences in behavior with differences in performance.
So that you're really going after the CEO level.
Is that right? So you're trying to attack time use at the CEO level and see how it translates down to management practices.
Can I just ask you about that?
It's fine. And so it seems like, what's fascinating in some larger senses, you're taking stuff that would have normally done by ethnographers like a time use survey, right?
Oh yeah. That was done by Meansberg initially.
And so, but you're really trying to scale it.
So for example, in the CEO thing, how are you implementing this?
How many organizations are we talking about?
We are talking about, about 1200 organizations in six countries right now.
Uh, yeah, but that's the concept.
I think that there is a lot that economists can learn from all these other methodologies and to the extent that you can find a way to scale methodologies that have value, but still deliver high quality data.
Why not? Yeah, no, absolutely.
And the second piece is again, something that I, you know, I'm motivated to do because of, you know, in part of my family history, my dad was a doctor, um, my brother, a hospital manager, in Italy, in Italy, in the same for some, at some point in the same hospital.
That's fantastic. Well, they were fighting.
So maybe. But you know, the, the, the deeper thing there is that, um, I do believe that part of the things that we measure for private companies, for example, the adoption of processes, standardization of processes can actually have a value also outside, um, you know, the normal firm, if you like, for example,
a hospital or a school where, you know, if you think about the complexity of production in this type of environments, it sort of makes sense to have some organizing principles that help coordination across functions or especially in times of urgency.
So what I noticed when I was, uh, you know, witnessing this discussions is that however, there is still a poor understanding of what management can really do in healthcare.
Um, and often is seen by clinicians as, as an imposition on their autonomy.
By the evil managers exactly.
There is no, it depends, but there is not a good translation.
Sure. So just to be clear, are the managerial processes thought to be the same?
So there's not a health specific set of things you're looking at or an education specific set of things you're looking at.
So the idea was to try and use the same ideas because they are so basic.
Yeah. And see how these ideas would translate across different sectors.
Yeah. So we, we did is we basically took the processes that we knew.
Had some, you know, interesting correlation with performance in manufacturing, and we ported the same, similar concepts to, uh, acute care hospitals and interviewed cardiologists, you know, cheap managers, again, middle managers in cardiology, orthopedics, and so forth.
We did 2000 hospitals so far in nine countries.
And again, the same basic method where you have the phone surveys and all this kind of stuff exactly the same methodology that is spectacular.
And the social returns obviously of getting management there, right are huge.
I, I do believe that there is a huge opportunity there.
Lots of people are, I'm not the only one saying it.
There's lots of fabulous research, uh, that talks about these things in the data.
We see that the hospitals, again, huge variation, even more than manufacturing.
Yeah. In some hospitals, I really think you have to, you know, pray God before you go, it's pretty bad.
In others, it's phenomenal.
Yeah. But for example, you know, it correlates with mortality rates of, you know, from heart attacks.
Yeah. So I really think there is an opportunity, you know, imagine if this correlation is causal, which we still have to prove, um, you know, there is a strong opportunity there to really have an impact.
I'm not just thinking of the medical errors work, which is, they're so important, right?
And this, this would be a clear way you could try to control.
So I'm, I'm curious if you kind of think back on this kind of decade of work.
What are some of the most memorable moments from all these conversations and what are the, what is the biggest surprise you think you've, you've taken away from it all?
You know, I would say at the time when we were running this management interviews, it was just phenomenal to see how much people wanted to talk about their experiences, middle managers, you know, typically nobody goes after them.
And sometimes we had these conversations, conversations that lasted for hours.
We had, you know, marriage proposals being made on the phone.
Oh my gosh. That's good.
But I think it was really interesting for me to, you know, being exposed to, as an academic to things that I didn't necessarily, I had not experienced personally, but still seeing, you know, how much motivation, how much passion these people were putting in their work and sort of listening to their
experiences was really, I think something that I don't know, but more generally, being part of a team.
You know, we were just digging in an area that perhaps had not been explored at that time and just being a part of a research team where you were discovering new things.
Well, I want to ask you about that because you and Nick and John have had this incredible partnership.
Yes. What's the secret to that?
They're great. Well said.
Well said. Now that's, it's really quite remarkable, right?
You've done a lot of work together by now and it's been enormously fruitful.
That's a really special thing.
You know, I think I've been very lucky to be honest.
Well, I'm sure it's more than that.
But, um, well, tell us, you, you mentioned your father and your brother.
Tell us a little bit about how you found your way to economics and to HBS.
Oh yeah. Well, to economics, quite boring.
I knew I wanted to do a PhD in economics since I was 12.
Something like that.
I know it's terrible, but that's really, I don't know.
I was really fascinated by economic history.
So you were at that age, you were kind of exposed to it and you were thinking about it?
Well, I think through my brother, primarily he's six years older than me and he's always been, you know, I'm a, I'm a copy of my brother essentially.
So he wanted to, he, he used to talk about economic history.
I had some good teachers in high school, got just, you know, the industrial revolution.
I don't know. These things.
Where did you grow up?
In Rome. In Rome. In the center of Rome.
Fantastic. I think also part of the story is my family, you know, they had a firm.
They built it twice.
Once my mom was from Libya and then at some point their factory, they had a big factory there.
They had to escape Libya and everything was ruined.
And my grandfather rebuilt everything.
In Libya or in Italy?
They came to Italy as refugees and they rebuilt everything.
So I was actually very close to the world of business and I experienced some of the dynamics of family firms very, very closely.
And then I guess, you know, I was lucky to be at a time in which you could look at business problems with large data sets.
And so on the one hand, very attracted by empirical work, statistics and large data, right at the time in which that data started to be available.
Yeah, you're claiming luck again.
I don't buy it. Very good.
Well, that was fantastic.
And I, and I, again, the article in HBR from last year is called, why do we undervalue competent management?
I think it's got a bunch of links in there to all the underlying work.
So you can, it's a nice introduction to all of the underlying work.
That's correct. And as I mentioned to you, Rafaela, when we were talking about doing this, we like to end these with a recommendation.
Yes. And so it's an after hours tradition.
So tell us about your latest recommendation.
It can be anything.
What's your recommendation?
So I'd like to recommend a book, which is the biography of Edith Penrose.
For those of you who don't know who she is, she's actually the mother of something called the resource based view of the firm.
And part of the story that she, you know, she lived, she started her career as an economist in the forties.
Part of her story was that management and people inside organizations were really a resource that could create growth.
And, and I'm just fascinated by the history of this woman who had four kids, started a field by herself, was managing everything.
I don't think she had a nanny.
And ended up creating a school of thought at a time in which there weren't many female women economists today in my MBA course, which is corporate strategy.
I teach her stuff. And it's just, you know, finally reading her life was a great experience.
That is really inspiring.
As are you. Well, thank you very much, Rafael.
That was absolutely spectacular.
The work again is why do we undervalue competent management?
It's available on the HBR site.
And it's really a fantastic pleasure to learn more about your work.
Thank you so much.
