Freakonomics Radio - 314. What Does a C.E.O. Actually Do?

Episode Date: January 18, 2018

They're paid a fortune — but for what, exactly? What makes a good C.E.O. — and how can you even tell? Is "leadership science" a real thing — or just airport-bookstore mumbo jumbo? We put these questions to Mark Zuckerberg, Richard Branson, Indra Nooyi, Satya Nadella, Jack Welch, Ray Dalio, Carol Bartz, David Rubenstein, and Ellen Pao. (Part 1 of a special series, "The Secret Life of C.E.O.'s.")

So I have one gripe about chip bags, especially for the single serve. It seems to me very suboptimal to have the only opening at the top, where you kind of have to jam your hand in there, and you can't really see what you're getting. Wouldn't it be better to have a bag that just laid out all the chips on the package, like a nice picnic blanket or something? Have you ever thought about that? But imagine that you're walking around eating a bag of chips, which a lot of people do. How would that work? If you open out the lase too much and a gust of wind comes by, you're going to see a lot of chips flying. We have to worry about all these little practical things. That's Indra Nooyi. Good morning. Nooyi is the CEO of PepsiCo.

So when it comes to little practical things to worry about, she has a lot of them. PepsiCo has more than 260,000 global employees and over 100 brands and trademarks, including 22 that each do at least a billion dollars a year in sales. Those include Pepsi, of course, and Gatorade and Tropicana and Quaker Foods. There are also seven separate billion-dollar chip brands, Lay's, Ruffles, Doritos, Tostitos, Cheetos, Fritos, and Walkers. So I'm out in the marketplace almost every week

looking to see how our products look like on the shelf, not from a CEO lens, more from a consumer lens. Because at the end of the day, I'm not just a CEO, I'm also a consumer. I'm a shopkeeper. I'm a gatekeeper of my family. So I look at our business through a different lens. And then I come back and I talk to my people about what I saw was good and what wasn't really good. Do you find it surprising that the CEO of a $170 billion company micromanages her potato chips? What about the CEO of an even bigger company, one built on silicon chips? I'm talking about Microsoft CEO Satya Nadella.

These hard decisions around what to pick and focus on is something that I believe a CEO uniquely has to do. That's not something that you can delegate. And what if you're the CEO of a social networking company with 2 billion global users? How micro do you think Facebook CEO Mark Zuckerberg gets? So when I was in Ohio, I sat down with a group of heroin addicts. And, you know, one of the things that was really interesting is, you know, when you're going through recovery, the first thing you have to do is detox, of course. But then after that,

the next thing you have to do is basically get new friends, right? And it turns out if you remain friends with anyone who you were using with before, then you are very likely to end up back using heroin and endangering your life. Okay, be honest with me. What's the first thing that comes to mind when I say CEO? For many people, the image is a caricature, either super villain or superhero. At the very least, you probably envision a luxe life with 100 rounds of golf a year. I think that CEOs work harder today than ever before. That's Jeff Sonnenfeld from the Yale School of Management. He spent more than 30 years scrutinizing CEOs.

These aren't people that are hanging out and golfing and running around country clubs and sipping on their sherrys late afternoon. These are people that are working nonstop, 40-hour days, eight days a week. Today on Freakonomics Radio, we are launching The Secret Life of CEOs, a special series that will get inside the minds of these rare and rarefied creatures. Some of the questions we'll be asking, what do CEOs actually do? What makes a good CEO? And how can you even tell? Why do CEOs make so much money? And are they worth it? How did they get to be where they are? And is it lonely at the top? You'll hear from lots of

big-time CEOs, as well as the academics who know them best. Indra is an unbelievably brilliant person. There's a peripatetic quality, of course, to Mark Zuckerberg. Ray Dalio is a paradox. Gerald Bartz is incredible. Nadella had kind of the toughest combination of all. With Branson, somehow the whole focus is on him. The protégés of Jack Welch, some of them loved him, some of them didn't. And if you find yourself thinking, hey, I'd like to do that too.

It's frankly a horrible job. You know, I wouldn't want it. From WNYC Studios, this is Freakonomics Radio, the podcast that explores the hidden side of everything. Here's your host, Stephen Dovner. Let's begin with this guy. I'm Nicholas Bloom. I'm a professor of economics at Stanford University. And if I were to just ask you, what's your general specialty? I work on trying to understand management practices.

So why some firms are better managed than others and how that helps improve their performance. All right. We are doing a multi-part series on CEOs. So let me ask you an incredibly rudimentary question. What does a CEO actually do? Because, you know, I think everybody knows what a CEO is, but I would argue that most of us really have almost no idea what a CEO actually does, both on a day-to-day and on a year-to-year basis. Their mandate is they're part public figures. So if anything goes wrong, all right with the company, they're the man or woman standing up in the press,

taking the flack or the praise. They spend about a day a week talking to Wall Street and other big investors. They also spend a huge amount of time on personal management, human resources. So generally they'll have something like five to 10 direct reports. People they'll see once a week that work, you know, the chief of finance, the chief of HR, the chief of information. It's much more, uh, you know, the, uh, the coach, if you think of a football team, it's very much like that with CEOs. They, you know, in some ways they're the masterminds behind the scene, but they don't actually, you know, throw the ball or actually have their hands directly involved in the business. Most people, when they hear the term CEO,

probably think of big firms like PepsiCo and Microsoft and Facebook and so on. But just to be clear, the median CEO in America, at least, has probably, what, four or five employees? It's an amazing fact when people mention CEO, they think of an older white man in charge of a company of 10,000 people. In fact, there are six million companies in America. The median company, so the 50th percentile, has three employees and the most common company size has one. So actually, almost every CEO out there you're going to meet is going to be in charge of five or 10 people. I deal with people who are going to be CEOs one day. That is Raffaella Sadoon. She's an economist who teaches at Harvard

Business School. I'm very interested in understanding how management and managers affect firm performance. So how much time does a CEO spend on their own thinking blue sky thoughts? There is very little time that goes into thinking alone or being in their solitary office and looking outside the window. We tend to think about the CEO sitting in an ivory tower, deciding what the organization will do, and then boom, they make a decision and the decision just happens and everybody's happy. But there is so much variation in how well firms adopt even the very basics of management. And you can't ignore it. That's not below the pay grade of the CEO. It's very important that the CEO keeps in mind both the strategic and the operational aspects of their jobs.

And yet, Sadoon says, that view is not shared by her MBA students, her future CEOs. They see strategy as paramount, how a firm responds to rivals or a shift in the market or consumer preferences. Operations, meanwhile, how goods or services actually get made and delivered, how the quality is controlled, how you monitor and track employee performance, that is not what they're thinking about. When we were teaching cases that had a lot of these operational aspects in it, my students were telling me that that's the easy stuff, right? That was not, you know, that's not strategy. That's easy. Everybody can do it or you can pay somebody to do it. And when you look at what happens inside firms, you just

know that that's not true. Raphael Sadun, along with Nicholas Bloom and the MIT economist John Van Rienen, analyzed data from more than 12,000 companies to try to learn what makes some better than others. Management practices, they wrote, can account for a large fraction of performance differences. And yet, they argued, quote, achieving operational excellence is still a massive challenge for many organizations. You know, the hygiene of management. So things like, do you collect data? Do you use it to analyze what's going on? Do you have thorough performance reviews? And in fact, many of the most successful companies in the world, companies like GE, McDonald's, and Walmart are excessively focused on managerial competence. You know, they're unbelievably detail-orientated, and that's one of their big ingredients to their success. So my sense of teaching our students is often they get overexcited about the big picture, sexy stuff of, you know, long-term strategy and kind of skip over the small details, which turn out to be critically important.

So how difficult is basic managerial competence and what keeps people from achieving it? You'd think that basic managerial competence is actually easy to do. I mean, in some senses, it's been around for over 100 years. So Frederick Winslow Taylor came out with something called scientific management in 1913, which was, you know, about using stopwatches and monitoring and data and, you know, roll us forward 100 years and it's been called in many ways big data. On the other hand, we just see tremendous variations. And I think it's because it's important, but it's boring and tedious and takes effort. And it's hard for people to get it right. And certainly in our data, we see tremendous variation in managerial competence. And I should say, not just in the private sector, you see the same in schools. You know, every parent's had experience with the good and bad teachers in hospitals and healthcare clinics, really just across the entire economy.

Okay, so basic managerial competence may be unsexy. Still, if it's so important, why don't CEOs pay it more attention? Here's one explanation. A good CEO has to play to five or six different constituencies. That, again, is Jeff Sonnenfeld of the Yale School of Management. He's also president of the Chief Executive Leadership Institute at Yale. There is a shareholder as the actual financial owners of the business. We have others with a stake in the business, the bondholders, debt holders. There are, of course, concerns as an employer that they're seen as an employer of choice. And of course, the employees have an opportunity that they're investing in training. And is there metrics that are created to take a look at their community

impact both environmentally and as a good corporate citizen to their immediate communities. But a fifth constituency is consumers. Are they getting healthy, a safe product? Is there a sense of accountability? Is there a warranty that the products work as they should and that companies are responsive? Okay, those are a lot of constituencies to balance. Being CEO is starting to sound like a fairly impossible job,

and I'd assume a fairly important one as well. I'd assume that who the CEO is at a given moment matters a great deal to whether that company will thrive. But let's not assume. Let me ask you this question not about yourself as CEO, but about CEOs of other firms. How much does the CEO really matter to the firm, and how can you tell? I believe a CEO matters a lot more than I probably thought before. That's David Rubenstein, co-founder and, until recently, co-CEO of The Carlyle Group, one of the biggest private equity firms in the world.

Private equity is a phrase that is used to explain the investing of money, typically in a company that is privately owned, i.e. it's not public. And you spend three to five years improving the company, incenting the managers to work harder, do more efficient things, and ultimately, after three or five years, you sell or otherwise liquefy the investment. Since its founding in 1987, Carlyle has purchased or invested in nearly 600 companies and sold off more than 300. Today, it has 271 companies under its control. So, besides serving as a CEO of Carlyle, Rubenstein was also involved in making sure that Carlyle-acquired companies have the right CEO,

and if not, bring in a new one. In all the companies Carlyle's invested in, I think the CEO has made the most amount of difference. I think the price we paid is probably the second most, and the quality of the company we invested in was probably the third most. So, if you told me you had a reasonably good company, a terrible CEO, I wouldn't invest in it. If you told me you had a reasonably good company and a great CEO, I certainly would invest in it.

Rubenstein plainly has massive experience in shopping for and observing CEOs. So who are we to challenge his opinion that the CEO is the single most important component of a firm's success? But here's the thing. It is largely his opinion. It isn't empirical evidence. That's what researchers like Nicholas Bloom try to come up with.

So I've been working for 20 years trying to understand what makes some firms perform better than others. To that end, Bloom has some good news and some bad news. The good news is that researchers have found, as David Rubenstein argues, that the leader of a company matters a great deal. The bad news? No one could really give us a straight answer on what defined a good or a bad leader. In other words, no one can really say what indicates or predicts or produces a good leader. And I must have pitched this question, you know,

dozens and dozens of times. You know, you look at the data and there's 10 different recipes for success. Maybe they each work for a particular case studies, but I've still 20 years later struggled to find anything that's the secret recipe beyond saying, sure, there are some people that are better than others, but it's damn hard to tell what it is. So on the one hand, I could say I admire your humility and I admire your candor. On the other hand, I could say, well, what good are you people then if you can't figure it out? Well, why don't I give you a great anecdote for sport? So, you know, imagine this. Imagine I took a thousand men and I ask you to pick which one of them would be best at tennis. Now, some of that's pretty easy. Professional tennis players tend to be taller than average. They're pretty athletic looking and lean. So I take the thousand, I whittle it down to the 50 taller than average kind of late 20s lean looking individuals. But beyond that, it's almost impossible to know unless you've seen how they've played before. So, you know, think about it practically, Roger Federer,

Nadal, Andy Murray, they look identical to just about every other professional tennis player. And so it is with leadership. By the time you get to people with basically, you know, great education, charming individuals, you know, some kind of prior experience, they're all in the frame for leadership. And sure, we can rule out people that, you know, typically dropped out of school at 15. But by the time you get to the subset of educated experience, you know, right background, at that point, it becomes incredibly hard to tell who's going to succeed. So one big problem with your kind of research is that you can't randomly assign CEOs to control groups and treatment groups. But if you could, if you somehow had the permission and resources to set up the perfect randomized experiment to learn whatever you want to learn about CEOs, or maybe it's several things, how to build the perfect CEO or how to preemptively tell a great one from a terrible one.

What would those experiments look like? Oh, you know, this is the researcher's dream is to go out there and randomly move CEOs around companies. And, you know, it doesn't happen in America. It probably happens in North Korea, but they're not sharing their data with us. But, yeah, ideally, you take a bunch of firms and a bunch of CEOs, and every five years, you flip coins and you rotate them around. And pretty soon, you'd find what makes a good leader and what doesn't.

The problem is right now, that just never happens. We have a lot of anecdotal evidence about CEOs, about what they actually do. That, again, is Raffaella Sidoun. But very little large-scale representative evidence. So coming up on Freakonomics Radio, let's get some large-scale representative evidence. So she's managed to persuade CEOs incredibly to release their time diaries on the confidentiality, what they do every half an hour for a week. It's coming up right after this. The most controversial fact about CEOs, at least in the mind of the public, is that they are just paid too much.

In 2016, the CEOs of the top 350 U.S. firms earned an average of $15.6 million, adjusted for inflation. That's a more than 900% increase since 1978. During that same period, the typical employee's compensation rose just over 10%. We've been hearing that good leadership is essential to good business performance, so maybe CEOs deserve that huge compensation. Why would a board pay it if they weren't getting their money's worth? And what's the right amount to pay a CEO? I think it's actually impossible to say what's the right amount a CEO should get paid. That, again, is the Stanford economist Nicholas Bloom.

The numbers they get paid now seem astronomical. I mean, you know, there are CEOs getting paid hundreds of millions. So I find it hard to defend those amounts. But I also am aware that there is no one that can definitively say that's too much or too little. The second point is you do definitely want to pay these individuals a lot. Why? One is you want to select people into the job. It's frankly a horrible job. You know, I wouldn't want it. Being a CEO of a big company is, you know, a hundred hour a week job. It consumes your life. It consumes your weekend. It's super stressful. Sure, there's enormous perks, but it's also, you know, all encompassing. And particularly for people with kids, you really want to, you know, make sure that they're motivated and also rewarded. And since

these individuals are running massive companies, their actions affect all of us, you know, make sure that they're motivated and also rewarded. And since these individuals are running massive companies, their actions affect all of us, you know, every day. So it's not just, you know, big publicistic companies. You've got to think of CEOs as hospitals and school districts and the government. So everything we do is affected by their actions. All that said and done, they're still way overpaid. That's Jeff Sonnenfeld, the leadership expert from Yale. And more than way overpaid. That's Jeff Sonnenfeld, the leadership expert from Yale. And more than being overpaid,

there's very little correspondence between performance and pay. There are some very strong companies historically, say United Parcel Service, which rank at the bottom in terms of compensation. But Philippe Dumont, who ran Viacom until recently, was being paid more than the very high-performing CEOs of Disney and Time Warner combined. And Viacom was a disaster.

People like to talk about, you know, what's the ratio against the average employee. No, even worse is the lack of correspondence between pay and performance. There is a lot of data to back up Sonnenfeld's claim. A huge study in the Journal of Management in 2000 found that the size of a firm accounts for more than 40% of the variance

between CEO salaries, while firms' performance accounts for less than 5%. A more recent study of over 400 large-cap U.S. firms found that higher CEO pay was correlated with worse than median stock market returns. So let's say you're trying to do better. Let's say you are trying to link compensation to performance. There are the standard metrics of share price, profits, and so on. But how can you tell that a different CEO wouldn't have accomplished the same or maybe even better? Which traits of a CEO could you measure to tell how good they are?

And are there real empirical and statistical indicators of what makes a great CEO? Or is it mostly just leadership mumbo jumbo? There's been over three quarters of a century of research on leadership traits. The bad news there is the sum total of it is very little of it matters in terms of static human traits that predict leadership greatness. A 2012 study found that American firms spend about $14 billion a year on leadership training. Every year, publishers put out approximately one gazillion leadership books, which are primarily used to wallpaper the bookshops and airports. Divulging the secrets of great leadership is, of course, a longstanding tradition among its most enduring practitioners, Sun Tzu, Plato, and, of course, Machiavelli.

These days, leadership courses and institutes are pretty much everywhere. So how good is the evidence to back up what they teach? As Jeff Sonnenfeld was telling us, the search for such evidence has been problematic for a while. In the aftermath of the Second World War, a flight of scholars who escaped the Holocaust came to the U.S. to help create something between sociology and psychology called social psychology

that really had no history before that. And its academic definition was really trying to understand the psychology of energized groups. But because of the experience so many had with charismatic leaders out of control, it became the vogue to focus on groups to the detriment of looking at leadership. We were afraid to look at charismatic leadership and often had some very simplistic notions out there in popular media, but in university worlds, we actually weren't teaching leadership even at the Harvard Business School. The supposed West Point of capitalism was very hard to actually take a look at individual leaders,

and that became the norm throughout business school. One of the first people to look at what goes into the activities of a manager is Harry Mintzberg. Raphael Sedun again. Who did some pioneering work in the 70s by measuring the activity of five CEOs, and that's what he based his PhD thesis on.

Henry Mintzberg's thesis was called The Manager at Work, Determining His Activities, Roles, and Programs by Structured Observation. Mintzberg got his PhD at MIT, and all the CEOs he studied ran firms in the Boston area, except for one, the Bull of a Watch Company in New York. The rest were Massachusetts General

Hospital, the Newton School System, Arthur D. Little Consulting, and a technology firm called EG&G, which, for what it's worth, was bought decades later by the Carlyle Group. Mintzberg shadowed each of the five CEOs full-time for a week. He recorded every activity they did during that time, every meeting, every phone call, and so on. Also, what they said and wrote and to whom, and what was said and written to them, and the action it elicited in the CEO. And there is a ton of knowledge that has been generated by his research. Mintzberg argued, for instance, that a CEO had 10 primary roles that fell into three categories,

interpersonal, informational, and decisional. For instance, among the decisional roles were negotiator and resource allocator, but also disturbance handler. Now, granted, it wasn't a large study, only five CEOs, remember, but at least Mintzberg started to bring some scientific method to the supposed science of management. But as Rafael Siddoun tells us, it didn't really catch on. After this initial push, there was very little large-scale representative research. And with very little large-scale representative research, a lot of the modern leadership gospel was based on observing high-profile individual CEOs. What we know about CEOs was pretty much based on exemplary people, people that were somehow famous.

And so it's hard to really think, you know, is that representative or not? Probably not. And you can imagine all kinds of faulty conclusions based on such unrepresentative observation. There's a long-held view that CEOs maybe are selected for being risk-takers. So imagine there are two types of CEOs. There's the boring ones and the ones that take huge gambles. At the beginning of their career, the boring ones plod along, but they never make it to the top. Of the huge gamblers, some of them win early on. They get promoted. They gamble again. Some of those win again, and they get promoted, and they gamble

again. And at the end of it, you can see that of the big gamblers, very few make it, but those that do get to the top. And so being a CEO, I think, is horribly selected on having taken a lot of big bets in the past and then successively paying off. So risk-taking is an attribute we end up selecting in our successful CEOs, but I'm not sure it's actually one that you'd want if you pick someone at the outset. Steve Jobs famously had his reality distortion field and ignored all his advice and pushed ahead and did well anyway. But I could mention Elizabeth Holmes at Theranos, who also apparently had a reality distortion field. And, you know, the company crashed and burned. And I'm sure there are 50 other examples of Theranos out there. So I'm very nervous about taking these anecdotes, looking only at winners. I think if you want to

do well, you want to look at winners and losers. Basically, take everyone that starts and see who tracks. And I think then you'll find the risk-taking CEOs, many of them crash and burn. We just never notice them because they drop out of the press. So that's the problem. What's the solution? Enter the World Management Survey.

The World Management Survey was a project started up back in the early 2000s to try and rigorously measure management practices in large samples of firms from around the world. So by now, 15 years later, we have about 40,000 companies from around 40 countries. So we have the US, the UK, but also Vietnam, Brazil, Australia, China, India. And it was just to try and be more scientific about what defines good or bad management. So we wanted to randomly pick companies so we have losers and winners. The idea being that if you had enough data and the data were truly representational, you could identify the specific qualities that contribute

to good leadership. And we just really struggled to come up with anything. We didn't find we could identify the secret source. But Bloom points to current research by Rafaela Sadun as potentially groundbreaking. She's been doing a bunch of studies. She's been doing a bunch of studies, not looking at who CEOs are, but what they do. So she's managed to persuade CEOs incredibly to release their time diaries on the confidentiality, what they do every half an hour for a week. That's right. Rafaela Sadun decided to pull a Henry Mintzberg. But we could also put Mintzberg on steroids. While the Mintzberg study relied on his shadowing five CEOs,

Sadun and her colleagues have been studying more than a thousand CEOs of manufacturing firms across six countries. Doing all this in person, as Mintzberg did, would have been impossible. So instead, what we did is we decided to create a project called the Executive Time Use Project, which essentially allows us to shadow not the CEO personally, but the agenda of the CEO. So we get information on every activity that happens in the life of the CEO. The researchers created from scratch

a call center and hired 45 people to call the CEOs or their personal assistants every day. In the morning and at night, to get a sense of what was planned in their agenda and what they actually did. We decided to also keep track of, is it a meeting, not only the type of meeting or type of activity, but also who was involved, the number of participants, who the participants were, whether the meeting was planned or not planned. Is the CEO maybe on the golf course? Is it a personal activity? Believe me, we have a lot of personal activities that are coded into our data. We don't use it for research, but it's kind of fun and validates the data a little bit. Validating the data is, of course, important.

After all, they're mostly self-reported data. And we all know that self-reported data have a habit of making ourselves look slightly better than we are. But Sadoon did have something to offer the CEOs in exchange for honest data. We told them, look, if you allow us to shadow what you do and to measure what you do, we will give you a report back that tells you what goes into your day, how long do you work, what do you spend your time on, and also will give you a sense of what your peers do. And that was the key because there is a tremendous appetite to know what am I doing? Am I doing it right? What are my colleagues doing?

Another potential limitation of the study, these are only manufacturing firms, so the findings wouldn't necessarily be universal. Also, the CEOs who chose to participate were slightly more likely to run smaller firms. Still, this time use data has led to some significant insights. What we find in the data is essentially two very different ways of allocating time. And in the paper, we label these leaders and managers. So let me start with managers. This manager is a CEO that spends a lot of time basically doing a lot of operational activities and often in meetings that involve one person at a time. The other type, which we call the leader, is a completely different way of allocating time. Instead of focusing on just one function, the meetings typically involve many functions at the same time, more people. Okay, so manager CEOs and leader CEOs operate quite differently.

What are the consequences of those differences? Sudun and her colleagues found that the firms run by a leader-type CEO were more productive and more profitable. But wait a minute. Earlier, Sudun and Bloom told us that CEOs who focus on good solid management, on operations, for instance, that they're more successful. And now, Sudun is telling us that a leader-type CEO

is more likely to be correlated with success than a manager-type. How can that be? We don't think that this correlation reflects the fact that leaders are always better for firms. What's that mean? If the firms run by the leader-type CEOs are more profitable and more productive, doesn't that mean that leader-type CEOs are more effective? We think that what's

happening in the background is something a little bit different, which is that leaders might be good in certain situations, might not be good in others, but that some firms that really need a leader actually end up with a manager. So it's more of an assignment problem of who gets hired to do what relative to a difference in the intrinsic quality of managers. In other words, the right CEO for a given company is, well, it's hard to say. It might be a manager type, might be a leader type. It might be a big, brassy, back-slapping, rally-the-troops type of CEO, or it might be a quiet, studious, lead-by-example kind of CEO. They're so different from outgoing to quiet to internal to external focus, just a whole range of practices. Nicholas Bloom's obsession to identify the secrets to good leadership has stretched on for years. But we just couldn't pick up any characteristic.

Maybe it just doesn't exist. Maybe there's just many different ways to be a great leader. Case in point, remember when Bloom told us the one thing that excluded someone from becoming a big-time CEO? And sure, we can rule out people that typically dropped out of school at 15. So how can you explain this guy? I've never had to report to anybody since I was,

well, since I left school at 15. And that is? My name is Richard Branson, and what do I do? I do everything Virgin. Over the next several episodes, we'll be trying to learn as much as we can about who CEOs actually are and what they actually do. As we've been hearing today, academic research on the subject indicates there's no template, no set of neat, identifiable characteristics that predict who will succeed and who will fail. We accept that fact. Indeed, we'll revel in it.

We'll talk to a bunch of CEOs and try to understand their lives, their decisions, their successes and failures. We'll talk about whether the kind of person who's capable of starting a successful firm is also the kind of person who's good at running it. We'll find out how CEOs get picked in the first place, and we'll take a look at the pipeline,

try to figure out why barely 6% of the Fortune 500 CEOs are women. We'll look at how CEOs deal with crises and make hard decisions, and mostly, we'll hear their stories. Yeah, well, I never started this to build a company. Mark Zuckerberg tells us what he did have in mind for Facebook. Ten years ago, you know, I was just trying to help connect people at colleges and a few schools. Mark Zuckerberg tells us what he did have in mind for Facebook. PepsiCo's Indra Nooyi and Microsoft's Satya Nadella talk about trying to be a CEO and a human being at the same time. It took me multiple years to even understand what had happened, because in some sense, I was more about, like, why did this happen to us? What happened to me? Well, I would say the first thing, and people don't talk about this a lot, but it is a very lonely job. You'll hear from Carol Bartz, former CEO of Yahoo, and Ray Dalio of Bridgewater Associates, one of the biggest hedge funds in the world.

The senior partner said that you're making people uncomfortable, you're demoralizing them with your straightforwardness. The Carlyle Group's David Rubenstein on making mistakes. Anybody that tells you they haven't made a mistake probably really isn't in the business or isn't being honest. Richard Branson will talk to us about founding versus managing. So I think too many young entrepreneurs just want to cling on to everything and they're not good delegators. Ellen Pao, former interim CEO of Reddit, talks about tradition versus modernity. Are we actually going to find that people are starting to hire people with different views and different backgrounds and different experiences who we really believe are going to change the system and not just perpetuate it with different players. And we'll hear from at least one CEO who seems to personify that status quo.

Treating everybody the same is ludicrous. And I don't buy it. It's not cruel and Darwinian and things like that that people like to call it. That's Jack Welch, former longtime CEO of General Electric. Some people consider him one of the best CEOs in modern history. A baseball team publishes every day the batting averages. And you don't see the 180 hitter getting all the money of all the raises. You don't win with a gang of mediocre players in business or in baseball. Coming up next week, how to become a CEO, origin stories, unlikely triumphs, and how to recover

from a really, really bad mistake. I blew the roof off the factory. Fortunately, no one was killed. That's next time on Freakonomics Radio. Freakonomics Radio is produced by WNYC Studios and Dubner Productions. This episode was produced by Max Miller. Our staff also includes Allison Hockenberry, Merritt Jacob, Greg Rosalski, Stephanie Tam, Vera Carruthers, Harry Huggins, and Brian Gutierrez. For this series, David Herman did the sound design and the music you hear throughout our episodes was composed by Luis Guerra. You can subscribe to Freakonomics Radio on Apple Podcasts or whichever podcast app you use. You should also check out our archive at

freeconomics.com, where you can stream or download every episode we've ever made. You can also read the transcripts and find links to the underlying research. We can also be found on Twitter, Facebook, or via email at radio at freeconomics.com. Thanks for listening.