Remarks by Steve Ballmer, CEO, Microsoft Corporation
Business Week Captains of Industry Series
New York, New York
November 13, 2003
STEVE SHEPARD (Editor in Chief, Business Week): It doesn’t matter how many times she says it, I still love to hear it. Thank you, Elaine, and thank you, everyone.
You know, when you think of Microsoft, you tend to think of Bill Gates. But Bill Gates is not running Microsoft on a day-to-day basis anymore. Three years ago he turned over the job of CEO to his long-time friend, our guest tonight, Steven A. Ballmer. Now, as Bill Gates focuses on what he loves best, which is software development, Steve Ballmer is shaping Microsoft in his own image, and that is what we will be talking about here tonight.
And those of you who have been here before are familiar with the format, after I finish introducing Steve, we’ll sit down in these comfortable chairs and have a conversation for about 45 minutes or so, and then we will take questions from you in the audience. So, please fill out the cards that have been given to you, and someone will come around to collect them, and we’ll try to get through as many of your questions as we can.
Now, just a little bit more about Steve Ballmer. He grew up in the Detroit suburbs. His father Fred was Swiss. He dropped out of university in Switzerland, but his language skills, he spoke six, landed him a job as a translator in the first Nuremberg Trial after the war. He made his way to America to Detroit, where he got a job at Ford Motor Company. Steve’s mother, Bea, was born in Detroit. Her father, Sam, had been a cobbler in Russia, immigrated to Detroit, where he owned an auto parts shop.
Steve, a promising student, won a scholarship to Detroit Country Day School, where he played football as a shotputter on the track team, and managed the basketball team. He also got a perfect 800 on the math SATs, nobody will be surprised, and was valedictorian of his class.
One of his crosstown friends in Detroit was none other than Scott McNealy, who is now the CEO of Sun Microsystems, and a rather outspoken critic of Microsoft. From the time Steve was 8 years old, his father told him he was destined for Harvard. And Harvard is where Steve and Bill Gates, another math whiz, were dorm mates in their freshman year. The two became friends. Bill left in his junior year to start Microsoft, but Steve stayed on to graduate magna cum laude in applied math, and set his sights on a career in business.
His first job was at Procter & Gamble, where he learned brand management working on Duncan Hines Brownie Mix. So, now we know what the secret is. He also shared a cubical and played Nerf basketball with Jeff Immelt, who is now CEO of General Electric. All along, Steve kept in touch with his college pal Bill.
OK, it’s now 1980, Microsoft is a small company, $12.5 million in revenue, and was running into problems. Steve had left Procter & Gamble, was finishing his first year at the Stanford Graduate School of Business, and was considering a summer job. He went up to Seattle and he talked to Bill Gates, and Microsoft co-founder Paul Allen, and they persuaded Steve to drop out of business school and join Microsoft. They agreed on a salary of $50,000 and some percentage of ownership in the company.
Steve was Microsoft employee number 24. One of his first acts was to tell Bill Gates that they needed to hire 17 more people, to which Bill responded, you’re trying to bankrupt me. Since then, Steve Ballmer has held virtually every management job at the company from chief recruiter to sales chief. He is also Microsoft’s chief cheerleader.
In Japan in 1991, he actually ripped out his vocal chords while rallying the troops and had to have surgery. He even popped out of a cake at Microsoft’s 25th anniversary. I don’t think he’ll do that tonight.
Steve met his wife Connie while at Microsoft, they have three sons, and the boys call Bill Gates “Uncle Bill.” These are relatively, I say relatively, tougher times for Microsoft. The years of 30-percent annual growth that the company experienced in the ’90s are probably over. Personal computer sales are slowing, and that means slower growth for Windows, and all the applications. Microsoft also faces a threat from Linux, an open-source operating system that is winning sales, especially among budget-conscious buyers. And need we mention viruses and worms that seem to target Microsoft in particular.
One of Steve’s tasks is to remaking Microsoft into a kinder, gentler company. Now, to its rivals making Microsoft kinder and gentler is like getting a tiger to become a vegetarian. We have a lot to talk about, so it’s time for me to stop talking up here and bring out Steve Ballmer.
Please welcome Steve Ballmer.
(Applause.)
STEVE BALLMER: You know more about me than me is what I concluded back stage.
STEVE SHEPARD: Wow, well, we do a little research, I hope most of it was accurate.
STEVE BALLMER: Pretty close, actually.
STEVE SHEPARD: Well, it’s an approximation, our software has bugs sometimes.
I’ve always wanted to ask you this, suppose that you hadn’t accepted Bill’s offer to join Microsoft, what do you think you would have ended up doing?
STEVE BALLMER: I actually was looking at three different things, consulting, which all MBAs at the time were looking at, investment banking, which the other group of MBAs were looking at, at the time. And I actually thought I would have probably taken a job as assistant to the president of a company called Progressive Insurance in Cleveland, Ohio for the summer. At the time it was a very small auto insurer, it’s probably fourth- or fifth-largest now in the United States, and it’s gone gangbusters. They were doing recruiting, and offering pretty cool jobs to MBAs.
STEVE SHEPARD: So you would have gone into the marketing end?
STEVE BALLMER: Into the marketing end.
STEVE SHEPARD: The course of history would have been entirely different.
STEVE BALLMER: I don’t know about that.
STEVE SHEPARD: Now, what it is about Procter & Gamble. You went there out of school, Jeff Immelt went there, Scott Cooke of Intuit went there. What was so special about people wanting to go to Procter & Gamble?
STEVE BALLMER: For undergraduates about my age, the thing to do, if you wanted to get kind of general management experience early in life, at a place where theoretically, and actually they taught well, was Procter & Gamble. It had that reputation, and as you point out, not only the folks you highlighted, but Meg Whitman was at Procter & Gamble about the same time I was. I don’t know if you mentioned Case was there, Steve Case. So a lot of folks went through there, and learned a heck of a lot.
STEVE SHEPARD: And you all ended up in technology, it’s just remarkable.
STEVE BALLMER: A bunch did, it is just an amazing deal.
STEVE SHEPARD: Otherwise you would have been doing brownie mix the rest of your life.
STEVE BALLMER: Either that or Moist & Easy Snack Cake mix, that was my other key product.
STEVE SHEPARD: The research didn’t pick that up, I don’t know. All right. So here you are now, and Bill decided he wanted to focus on software, and you became the CEO. Talk a little bit about the transition, what we heard, of course, is that it was not easy, there was a little bit of shouting back and forth. Describe the process of transition.
STEVE BALLMER: Yes. Bill and I have been friends for 29 years, and that’s been constant. We were friends in college, we stayed friends. When I first got to Microsoft, frankly, we had a difficult period, also. We’re friends, but a difficult period, because we had to kind of sort through, despite our friendship, exactly what the working dynamic would be. That probably took us about a year to kind of sort through. We were still friends, but husbands and wives are friends, too, and things are not always totally smooth sailing.
We’d been in a pattern of behavior in how we worked together for a number of years, and when Bill asked me to take over as CEO it was clear he really wanted me to be CEO. And I’d say, it was also fairly clear shortly thereafter that neither he nor I really understood what that meant, how were we really going to change our working dynamic. And it probably took us, again, almost a year to really sort that through. But, things are we now have a new kind of way of interacting with each other. I know when to defer to him, he knows when to defer to me. And we continue to work together in a super incredible fashion.
STEVE SHEPARD: Generally speaking, how do you divide the responsibility?
STEVE BALLMER: I think the best way to think about it is, Bill focuses in on the technology vision, key technology strategy, key business strategies that are deeply rooted in the technology. I put our intellectual property, patent stuff in that category, even some of the work on antitrust, frankly, winds up being more Bill’s focus than mine, because a lot of it is about product design. I do the rest, which means really managing the company, running the company, setting the business vision for the company. Those would be the things that I focus in on.
STEVE SHEPARD: And you can go pretty far on your own without talking to him?
STEVE BALLMER: I’d say and vice versa, in a sense. There’s a set of things that are clearly Bill’s expertise, and a set of things that are clearly my expertise, but we’re in big meetings all the time, and we sit down one-on-one every week. I’d say there’s probably Bill likes to say, he probably sends me over a typical weekend 10 or 15 pieces of e-mail, and I don’t respond to them until Sunday night. That’s my deal with my wife, and he gets the Monday morning set of replies back. So there’s a lot of dialogue on stuff that there needs to be dialogue on.
STEVE SHEPARD: How would you say you have evolved as a manager since becoming CEO?
STEVE BALLMER: It’s been amazing in a certain sense. You would have thought that after 20 years at Microsoft I kind of understood what the job was, and what I needed to do, and what I really figured out is the role of the CEO is quite different than I thought, both in terms of what he needs to focus in on, and in some senses what the organization needs or doesn’t need from you. And I probably thought I was going to have my fingers much more in every pot, as opposed to delegating as much as I feel like I have. Certainly, the external issues other than sales are far more important than I might have gauged, whether that’s corporate governance style issues, representation with government, et cetera. So there’s a set of things that I’ve really only come to appreciate since taking the CEO role.
STEVE SHEPARD: Did you find it hard to back off from the micromanaging part of it?
STEVE BALLMER: No, I had to find something new to micromanage, very truthfully. You’ve got to have something. Microsoft I don’t know what it’s like in a lot of companies, but you’ve got to have something at our place you really just totally know, you’re expert on, or people start losing respect for you. And for me it used to be this business, or that business, or the other business, and it’s evolved to the place where I’d say the thing I understand best now, and I micromanage in a sense, is the processes that we use to run the business. That’s the thing I think I control and organize, and run more than anything else.
STEVE SHEPARD: Give me an example.
STEVE BALLMER: We do about $6.8 billion of R & D. The whole of our R & D is supposed to be bigger than the sum of its parts. If you’ve got a Media Center PC in your living room, if you carry one of the Windows mobile phones with you, if you’ve got a PC, if you’ve got a laptop, those things should work together perfectly. And that means those can’t all be separate independent divisions. Bill’s got a vision for how that technology integrates. But what’s the management process that lets those things integrate, but still gives each of those groups the chance to be close to their customers, to respond to customer issues, have some financial authority and control, and autonomy? And we’ve been working very hard to try to get the management system to support Bill’s integrated innovation vision in place.
STEVE SHEPARD: Talk a little about this kinder, gentler stuff. What does that mean and is there, in your view, a kinder, gentler Microsoft?
STEVE BALLMER: I think everybody wants to sort of say it that way, so let me characterize how I think about things. When you have gone through the kind of experience that we went through over a multi-year period, with government authorities, et cetera, it really does cause you to step back and reflect, who are we, where are we, what are we doing, in addition to all of the additional as we settled our matters with the Department of Justice, all the additional responsibilities that we signed up to, but there’s more of a reflection. And it really got us to say to ourselves, look, no matter how we think about ourselves, we may think of ourselves still as young upstarts coming up, struggling every day, that’s not how the world thinks of us, that’s not how our customers think of us. The bar is in some senses, the expectation bar, be it from government, be it from customers, be it from industry partners, the bar is different, and the bar is higher, and we have to meet those new expectations. We can’t think of ourselves in exactly the same way.
So the balance we have to hit is, how do you really behave as responsible leader for the industry, and yet at the same time let the company know, the employees know, we’re still going to be innovative, we’re going to still try to provide great value, that means our innovation at low prices, we’re going to compete. If we lose a customer we’re going to feel bad, we’re going to want to know, what do we need to do better next time so we don’t lose that customer the next time. So how do you hit that balance between being forceful, and aggressive, and still being a great participant, talking to your industry partners, even if you compete with them sometimes, right level of cooperation, right level of cooperation with government. And we’ve really worked hard on what I would call that theme, which is responsible leadership, because our customers, and other involved parties do see us as industry leader.
STEVE SHEPARD: How far along would you say you are in that process, 30 percent, 50 percent, 70 percent?
STEVE BALLMER: It’s hard to pick a number, I’d say we’re farther than 20 and less far than 80. We’re someplace in the range. It takes a while to understand what does it mean to operationalize that. You can say it, there’s a lot of things I’ve said to our company that nothing happens, because we haven’t figured out how to make it concrete. People understand more what it means to make it concrete, what are positive things that we need to be doing as an industry leader with government. Take spam, where we’ve had a big initiative, where we’ve been trying to work with our industry, and with government on anti-spam initiatives. Take the work we’ve been trying to do to reach out to our competitors. We’re really partnering now on a set of important interoperability standards with IBM, despite the fact that IBM is in every sense our biggest competitor, a very different place for us to be. I think we’ve made appreciable progress, and yet I’d be the first to say, I know our customers, our industry want us to still do better.
STEVE SHEPARD: Do you think there’s a kinder, gentler Steve Ballmer?
STEVE BALLMER: Sure. Of course there is. You know, I’ve worked hard on it. And so in a sense, it’s almost harder to say than how far has Microsoft come. I’m still every bit as energetic, and able to get kind of pumped up about something as I ever have been. But just as I’ve had to understand what my role is in Microsoft as CEO, and it’s a little different, I also have to recognize that the role that I have to play externally as the leader of the company is different. I can bring all the excitement, and all the enthusiasm, and all the passion, and all of the desire to do the very best job we can for our customers, I can bring all of that, but I have to bring it in a different way.
STEVE SHEPARD: Does this kinder, gentler notion, however you want to define it, does that include befriending, reconciling with people like Scott McNealy, or Larry Ellison, or some of the other outspoken critics over the years?
STEVE BALLMER: Our industry in some senses has been unique, probably because it all grew up with a bunch of young people running it, and kind of a different kind of environment. But it’s always kind of unique to the degree to which there has been this kind of public feuding in our industry compared to others. And certainly we’ve reached out. We’ve reached out with IBM. I went down shortly after I became CEO to visit with Larry Ellison, and talk about how we work cooperatively on things that are important to our customers, their products working with our products. I hadn’t seen Scott in years and years, we played in the same golf tournament a year-and-a-half ago, and very amicable reacquaintance so to speak. It’s tough when you’re in a lawsuit with somebody, I have to say. It doesn’t sort of flow so naturally for me. But, yes, that is part of the deal is trying to figure out how do you resolve matters. And our new chief counsel has been very active on that, particularly on the government front, trying to resolve matters constructively.
STEVE SHEPARD: It’s very brave to play golf with Scott McNealy, he’s very good.
STEVE BALLMER: Yes. If you put him in the right environment with the right handicap, it can be quite helpful. He is very good.
STEVE SHEPARD: I was going to make the same point about the industry, it strikes me the same way, that there really I can’t think of another industry in which there’s been such public hostility among the major players, and all very successful, bright people, and they really go at each other. It’s remarkable.
STEVE BALLMER: Well, how much of it is real, and how much of it is the fact that it also makes good theater? We’ve never been that active on the good theater side of that. We’ve been more what I would call the defensive side of that. But, I think it’s a good maturation for the industry.
STEVE SHEPARD: Yes. Who are some of the other managers around, whether in the tech industry or elsewhere that you admire?
STEVE BALLMER: You never can really peer inside anybody else’s organization. It’s something I’ve learned. I went on my first non-Microsoft board a few years ago, and I thought I knew the company, a very good company, Accenture, I thought I knew them very well, and you don’t know what you don’t know until you get a chance to be a little bit inside. And so I admire the output or what I’ve read as opposed to what I can really see and know.
I certainly admire what the folks at Intel, Andy Grove and others have done, because I’ve had a chance maybe to see a little bit more inside that organization.
I’m fascinated by what Welch did and what Jeff is sort of reshaping at General Electric, because clearly it’s one of the most thoughtful places in the world about the process of management, so I’m interested there.
I’ve studied a lot what goes on in the auto industry and the pharmaceutical industry, because if you look at other industries with large R & D spend, people we should emulate, there are only a few industries to go look at, but I haven’t found any one company that I’d just point to or any one individual and say, yeah, that’s the guy.
STEVE SHEPARD: Talk about the antitrust thing in the sense of what have you learned from that whole experience.
STEVE BALLMER: Well, I think the number one learning is the thing I talked about earlier, which is the sense of increased expectations on the company from all parties and really stepping back and say it doesn’t matter how we may have seen ourselves; we have to see ourselves as others see us, and that by far is the biggest learning.
I mean, we have learned what the process looks like to actually work constructively with government to resolve matters. We know how hard it is and we’re working super hard to make sure we absolutely comply with all of the responsibilities and obligations that we pick up in our consent decree, and it’s harder probably than I guessed it would be, not to control errant behavior, you just want to make sure you absolutely have the right processes in place to make sure you’re doing exactly what you committed to do to the U.S. government.
STEVE SHEPARD: Now, the antitrust process is unfolding again in Europe right now. Your lawyers are over in Brussels yesterday, today, tomorrow I guess as well. What’s the status of that? That’s essentially the same kind of a case, what you can put into the operating system. Tell us about how that looks to you.
STEVE BALLMER: It’s some of the same kind of a case, some different kind of matters. Our great desire would be to try to work through a constructive way to sort of resolve the matters that we think will work for our customers and will work for the Commission.
The hearing that we’re involved in now is certainly a part of the process in Europe, but even while this is ongoing we look forward to a constructive dialogue that hopefully results in some kind of a reasonable settlement for all parties.
STEVE SHEPARD: The issue that they seemed to be most concerned about is the Media Player being built into the basic software.
STEVE BALLMER: It’s certainly a key thing on the list that the Commission has wondered about and had concerns about.
STEVE SHEPARD: Now, what happens if they say you can’t integrate the Media Player into the operating system?
STEVE BALLMER: Well, of course, we comply with government bodies appropriately. I think there’s still quite a process to work through. There’s this hearing, there’s more settlement discussion. Then the Commission has to really decide what it wants to do, because they haven’t made a decision but they’ve sort of expressed their views. There’s a whole big process in Europe that I’m not the technical expert in.
But, as I said, the number one hope we have is to try to work through constructively some kind of settlement.
STEVE SHEPARD: But this is core, fundamental stuff to Microsoft, because it’s product design and Bill I know feels very, very strongly and you do, too, so this goes to the heart of the company in a way if they say to you, “I don’t care, it’s coming out, we’re not going to let you sell that in Europe.”
STEVE BALLMER: I would say differently than, “It’s product design.” It’s doing what we’re supposed to do for our customers. We think our job is to take new ideas, new innovation and integrate them in a way that makes them super easy, super useful. We think that’s what customers actually appreciate in terms of value that we can add to the marketplace.
So sometimes you could say it’s all about sort of, we need the arrogance of our own product design. No, it’s about our fundamental value proposition, the fundamental way we serve customers essentially. That was all the process we went through with the U.S. government, a painful process, which probably, I don’t know, maybe in 20/20 hindsight there would have been a better way through it.
But the key thing for us is even under our settlement with the Department of Justice many new responsibilities, but we do continue to have the right and ability to serve our customers in the ways they expect, which I think is important and I think would be important in Europe, too.
STEVE SHEPARD: When does this come to a head? When do you get a decision? In the spring or —
STEVE BALLMER: The Commission works on its own time. According to newspaper reports, it would be the spring.
STEVE SHEPARD: Let’s talk about future growth for Microsoft. There’s some talk the stock is relatively flat, you’re starting to pay a dividend, you’re starting to act like a, quote, “mature company.” Growth rates aren’t as high; how could they be? What are the opportunities for growth? Where do you see the growth coming from the next couple years?
STEVE BALLMER: Yeah, let me frame it up in a vision sense, because that’s the right place. The question I ask our own employees when they might ask a question like that, which is different than when you might talk to a financial analyst, just do this thought exercise. Ten years from now will the world of technology be A) largely the same or B) largely evolved and different than it is today? I think the answer is clearly B, things will change, new value will be added, and that means there’s opportunity. Exactly how you convert that change, that dynamism, that additional value that can be delivered to businesses, to consumers, to small businesses, that’s not 100-percent clear.
I mean, we always have a plan and a strategy, but the fact that our industry is going to continue to increase the value that it adds to society, that creates opportunity. And whether that creates a 5 percent a year, a 10 percent a year or 15 percent a year opportunity for our company, I don’t know. I mean, I can make up a scenario that looks like a lot of those different numbers, not all of them but a lot of those different numbers and other numbers, but the fact of the matter is the opportunity is there so we’re investing the R & D behind the opportunity, $6.9 billion in R & D behind the opportunity to positively change the world.
And that’s how I think about what’s in front of us. And will it translate into this growth rate or that growth rate, I have some senses of what might happen but we make no predictions. We talk instead about the opportunity and our commitment to make sure that we’re the leading company in helping our customers realize those opportunities.
STEVE SHEPARD: But you’re operating in new areas now. Games, for example, was not in the original business plan I’m sure, MSN, information businesses. Talk about some of these new kinds of businesses.
STEVE BALLMER: What we are essentially, you could say, what we’re doing is we just continue to follow two basic sort of compasses. One compass is the compass of technology, what will technology permit. Given the fact that Moore’s Law is continuing, and it’s continuing not only on processor but disk and communications, what new scenarios do they enable, and new software advances? And then the other compass we follow is what other things do our customers want to do.
I mean, think about the PC, it’s an amazing morphing device. First it was a programmer’s tool, then it was sort of an accountant’s tool, then it became a secretary’s tool with word processing, then it became a games machine and then it became a learning machine, then it became a communications machine, then it became a browsing machine, then it became a reading machine, now it’s an entertainment machine, a music machine, et cetera, et cetera, et cetera, et cetera.
The range of things that you can do, so you look at, quote, “our new businesses,” in a sense they have a new P & L associated with them but in all cases pretty much we’re following technology and customer interest and sort of developing the right technologies and the right products to exploit those.
I believe within today probably 65, 70 percent of American households have a computer. I think if you go out 20 years from now 60 to 70 percent of American households will have at least two or three computers. But where will they be? What will they be doing? What are the scenarios? How do they need to connect to each other?
That’s what gets us into Xbox. We don’t just sit there and say, “Oh, nothing to do today; let’s find a new business to get into.”
STEVE SHEPARD: Let’s play games.
STEVE BALLMER: Yeah, let’s go play games. No. But if you think about it, the scenarios that open up really are amazing, whether it’s the way you keep track of your own digital videos and move them around the house, and there are so many things that are just going to be so different.
STEVE SHEPARD: The next-generation operating system, which you’ve talked about a little bit, codenamed “Longhorn,” what’s going to be dramatically different from the current system that you can tell us?
STEVE BALLMER: Yeah, and there are sort of two dimensions and I’ll stick to the second — three dimensions. One dimension is what’s different from the standpoint of people who take care of computers. Computers are still too hard to take care of. Whether you’re an individual at home or an IT shop in a company, they’re too hard to take care of. There’s fundamental advances to make them easier to take care of, to secure, to back up, to manage, et cetera.
Number two, developers want to write new programs. Sometimes new programs are hard to write without fundamental advances in the underlying system, because it’s just too expensive for company after company after company to do the technology. We have a load of advances that we think are fundamental to changing the way applications get written.
At the heart of it from an end-user perspective there’s probably three things I would highlight. Number one, it’s still too hard to pull information together from a variety of sources. I mean, if you really want to create your own personal page, homepage, I mean, you can kind of do that on MSN or Google, but you only get what MSN or Google want you to have. You can’t take, as I might want to, my wife’s calendar with my kids’ school calendar with my calendar with Microsoft sales data, boom, and just be looking at it and integrating it.
If you want to find things, it’s not that easy, it’s not that hard but people know how to search the Internet. Will Internet searching look different five years from now? You bet, and we’re trying to make advances in “Longhorn.”
But even the way you search your own information today, you’ll want to see advances. Just think about photos for a minute and so this searching and management of information is a big deal for “Longhorn.” Think about photos. If you actually had a library of a thousand digital photos, people say, “Well, no problem, I’ll go buy the Adobe product or this product or that product.” But if you then want to sort of correlate those photos with your calendar, which I do want to do, because I like to relate them to different vacations and blah, blah, blah, blah, blah, no way to do it, the system doesn’t make it easy for me as a user to kind of correlate and integrate a bunch of different information.
So managing information, searching for information, kind of integrating information from multiple sources, those are some of the key kind of end-user conceptual breakthroughs in “Longhorn.”
STEVE SHEPARD: When you say search capability in Longhorn, do you mean sort of like Google?
STEVE BALLMER: There are two parts to it. There’s the front-end that helps you sort of express your search and then there’s back-ends like Google and MSN and Yahoo and many others that have a certain way of characterizing data to put into a front-end, and I’m thinking more of the former than the latter.
STEVE SHEPARD: And what’s the timeframe on “Longhorn,” would you say?
STEVE BALLMER: When it’s done. (Laughter.) No, it’s the biggest advance of Windows that we’ve done, in some senses, for 10 years, for 15 years. It’s probably not that important whether it’s three months early or six months later. The most important thing is it really sets up these fundamental advances for the three audiences I described.
STEVE SHEPARD: Will it have a Windows name like Windows 2005 or whatever it is?
STEVE BALLMER: Yeah, it will have some Windows name. At least some of these magazines tell me Windows is a pretty valuable brand name. I don’t think we’d start using another one at this stage.
STEVE SHEPARD: All right, well, so we knew when Windows 95 came out, we knew when Windows 98 came out and we knew when Windows 2000 came out, so this will be Windows — you’re not going to tell me, all right.
STEVE BALLMER: No, I’m not going to tell you, but you knew that.
STEVE SHEPARD: Yes. (Laughter.)
Linux, how serious a threat?
STEVE BALLMER: We have good competition. It’s interesting competition, because it’s good in a different way than some others. We’ve always been able to go in, in any sales proposition, for as long as I’ve been at Microsoft and say we have the best product and we have the cheapest product. It’s a great way — I mean, anybody who’s in business has got to love that proposition. You can say we think we’re the best and you can say we think we’re the cheapest.
We have a competitor now that’s a weirdo competitor in a sense, because there’s no company behind it, you don’t exactly know who builds it, you can’t call somebody up and find out exactly where it’s going, so it’s very unusual and it’s free. I mean, we’ve got to have some businesspeople in the audience. Most of them aren’t competing with anything that is, quote, “free.”
Now, I say, “quote, ‘free'” because in anything you do with computers the software is a small part of the total amount of cost and energy and time you put in. So it’s an unusual competitor.
And in some senses I’d prefer to say, look, what we have here is a small price disadvantage. (Laughter.) It’s the first time we’ve had a price disadvantage. Most analysts think the price of Windows to our hardware customers, people like Dell, is about 50 bucks. So are they free? No. We’re $50 more expensive, somebody might say. If you stop and think about it, most people are going to own their PCs for four years. So do we offer $12 a year of value where you can run tremendously more applications, it’s tremendously easier to take care of, $12 a year? I mean, for people who are spending $90, $100 a month on cellphone bills and we’re talking about saving you hours and hours and hours of your time, I think it’s a pretty good value proposition myself. (Laughter.)
STEVE SHEPARD: And it won’t matter that somebody else is zero?
STEVE BALLMER: It’s $50 cheaper. (Laughter, applause.)
STEVE SHEPARD: That’s what’s known as applied math, now I know. (Laughter.)
All right, so why are people buying it?
STEVE BALLMER: Well, on PCs most people aren’t.
STEVE SHEPARD: Are not, yeah.
STEVE BALLMER: I mean, truthfully, it’s not that common.
STEVE SHEPARD: In servers.
STEVE BALLMER: On servers it is. On servers we’re about a $500 product and servers are used by multiple people. And when you reduce it down it’s about the same thing. It’s about $50 a head, 10 bucks a year.
And so the question is why, why would they use it on servers? And there are some scenarios. And you can sit there and get into philosophical, religious things, ooh, it’s about populism versus capitalism, it’s about blah versus blee, it’s about politics and government in the city of Munich, blah, blah, blah, because they picked Linux for something, what is it about.
The truth is there are some things where we just tell our engineers, look, they’re better than we are, right now they’re a little better than we are on a couple things. If you have an application that you’re trying to get off of one of Scott McNeely’s big old expensive pieces of hardware and you want to put down — (laughter) — OK, I couldn’t resist one small one. And you’re trying to get down onto a piece of PC hardware, Linux is more like UNIX. And there are cases where I’d tell you it’s easier to move an application from UNIX to Windows than it is UNIX to Linux, but that’s not the prevailing wisdom. So we tell our engineers, “Understand everything about it. You have got to be a better UNIX than Linux. Work it, work it, work it, work it, work it, work it.”
And that’s where we get our value proposition. Or you’ll find a case where somebody says there’s not enough innovation in this scenario from Microsoft to justify a $500 price point. And there are some cases where we do have lower price versions of the Windows servers that are cut down that are more appropriately priced for the scenario. So instead of getting caught up in philosophy and blah, blah, blah, you have to think about it like a business problem: How much value do you offer for your price premium? For us it’s just been new because we’ve never been at a price premium. We were cheaper than CP/M, our first big competitor. We were cheaper than Lotus, we were cheaper than WordPerfect, we were cheaper than Novell, we were cheaper than Sun, we were cheaper than Oracle, cheaper than IBM, and now you’ve got a guy basically you can’t be cheaper than. (Laughter.)
But there are plenty of companies out there that sell products at a bit of a premium and we’re just at a bit of a premium — in the grand scale of what people do with these things, we’re really just at a bit of a premium.
STEVE SHEPARD: Now, you said Linux is not really a company, but IBM, which you characterized earlier as your most important competitor, is getting behind Linux in a big way. So does that increase the threat?
STEVE BALLMER: Yes and no. I mean, in a sense because the thing doesn’t come from a company, having some PR air cover from a big brother company is a useful thing for Linux. And in a sense I think of our competition as almost more with Linux sometimes even than with IBM. Now, IBM is a company so you can fixate on that. It’s harder to fixate on just a product without a company behind it, but certainly that’s been beneficial.
How much behind Linux is IBM really? They’re doing a lot of advertising but the truth is IBM doesn’t deliver Linux. I mean, if you asked IBM to send — I picked on you because you were sitting right there — if you asked IBM to send you a copy of Linux they’d say, “We won’t do it.” They’d send you to see Novell or Red Hat or one of the Linux distributors. If you asked IBM what’s the future of Linux, they’ll have to say, “We don’t know, go talk to somebody who does,” because they don’t control it. If you asked IBM, “When are you going to get a fix for a problem in Linux,” they’d have to say, “I’m sorry, we’re not in charge of that.”
So I’m not saying they’re not behind it in a certain sense, but it’s not like IBM’s mainframe software where they stand behind it with an absolute commitment and an ability to execute on that commitment.
STEVE SHEPARD: I guess the fundamental question is this, that you had and have a monopoly in the PC operating system, and you’re not going to have that, you don’t have it and you’re not going to have it in the server business. Is that a fair statement?
STEVE BALLMER: I have to say nothing other than that’s a fair statement. We are trying to build share in the server market and, in fact, we did build share in the last 12 months. It’s an interesting thing. Our share of the server business is up in the last year and Linux’s share is up in the last year. So it’s the things that are not Windows server and are not Linux, like Sun’s Solaris, like Novell, like old versions of UNIX, those are the things that have been getting crowded out as we both have built share in the last year.
STEVE SHEPARD: Is Linux’s share in the server market bigger than yours now?
STEVE BALLMER: Not close to that. I think most analysts would say we’re probably someplace between 2.5 and 3 times the number of units per year that they have.
STEVE SHEPARD: For server.
STEVE BALLMER: For server.
STEVE SHEPARD: The viruses, the worms, a lot of blame came to Microsoft. What can you do about this? What are you doing about it?
STEVE BALLMER: There’s a lot we’re doing and a lot we have to do. The first thing everybody says is, why don’t you just make your software all better, and the truth is it’s a reasonable starting point for people to say. There are 500 million computers installed in the world though, and they’re not going to all go get better overnight. It’s not going to work that way. So, even on our new products, where we built new tools, we trained our people, we have new management approaches to try to ensure higher and higher levels of security, you know, we think we’re doing a pretty good job.
But I think it would be unwise for us or anybody to assume about any product that it will be perfect. The first thing you have to do is up the quality level. The second thing you have to do is, OK, if there is a problem, how do you help your customers fix that problem?
Just a small show of hands, how many people are running Windows XP on your computers at home? And how many of you would have Windows Update turned on, if that means something to you? And if you don’t have this thing called Windows Update turned on, I would recommend it because, it automatically will come and take care of your computer, you can set it up to install critical security patches. And we have to see how we bring we’re working now to bring that same kind of approach and technology to our corporate users that we have to our home users. So, remediation is an important step.
A third important step is, how do we build essentially layers around PCs and corporations that essentially protect them from the bad guys. And that’s one of the big learnings of some of these recent viruses, there’s some very good technology that we and others have that aren’t really fully and properly deployed. So we have to make it easier to deploy those technologies and kind of build little shields around things. There’s a feature in Windows XP called the Internet Connection Firewall. People who had it turned on, their PCs were fine. People who didn’t have it turned on, or hadn’t installed the patches, they weren’t fine. How do we make that all natural and normal? What’s the corporate equivalent to that? And we’ve got our engineers this is the number one area of R & D investment for us across the board, I would say.
STEVE SHEPARD: And how do you think of spam, the same way, we’ll never have a perfect system to keep it?
STEVE BALLMER: Probably never perfect, but so much better than we have today, and that will be a combination of factors, kind of like security, some of it will be technology, some of it will be kind of filters and business practices, and some of it will actually be cooperation of law enforcement. I should have said that in the security case, and I certainly say it in the spam case, unless bad guys are being punished, there’s not enough of a deterrent. I mean, if somebody bombs an empty building and nobody dies but there’s huge damage, everybody thinks that person is really not a very good person. If somebody spams you, or if somebody does hundreds of millions of dollars of damage by sending out a virus, people say, oh, they’re cute kids. They’re hackers. The truth is, you can do just as much or more damage nowadays in the world of the Internet as you can in the physical world, and I think it’s important that we and other companies in our industry really do the right partnership with law enforcement to make sure there’s enough deterrence out there.
STEVE SHEPARD: Is the technology piece of the solution that you’re talking about, is that a Microsoft
STEVE BALLMER: We’re certainly working on anti-spam. We just came out with our new version of Microsoft Office, Office 2003, the new version of Outlook, the e-mail package is in there. It has specific advances in its anti-spam capability, and we have a big investment in a lot of things that we think will improve that.
STEVE SHEPARD: This is something that everybody would be very happy to see Microsoft build into the operating system. You won’t get any antitrust complaints on that one.
STEVE BALLMER: Who knows? But, it’s a good example of an interesting question. Customers would like us to build that kind of stuff in, and I’m sure there’s some companies out there
STEVE SHEPARD: Who are building anti-spam things themselves, too.
STEVE BALLMER: Who might find it problematic. I don’t know what the government authorities would think, but it’s a tricky one. Trickier than in your question.
STEVE SHEPARD: I didn’t think of it that way, but that’s probably right. One more thing I want to ask you before we go to the questions from the audience, which has to do with China. Well, there are two questions related to China. One is, first of all, the immediate affect on Microsoft of the piracy of your software in China, how big a problem is that for you, and what can you do about that?
STEVE BALLMER: Well, most of our software used in China has been pirated. Overwhelmingly most. When you say how big of a problem it is, I might say how big of an opportunity is it. And it’s a big opportunity.
STEVE SHEPARD: What’s the lost revenue?
STEVE BALLMER: It’s many, many, many hundreds of millions of dollars of lost revenue, certainly many hundreds of millions of dollars.
STEVE SHEPARD: A year?
STEVE BALLMER: Per year.
STEVE SHEPARD: Per year.
STEVE BALLMER: So I view it as an opportunity. One of those big upsides that when you ask me about growth rates, I’m sitting here thinking to myself, only if. So how do we get there? What do we have to do more specifically in the Chinese market for the products, how do we have to think about different dynamics, what are we going to need to do in cooperation with government in China. How do we be a better citizen, and contribute more to the development of the Chinese software industry, so the Chinese local companies care about the piracy issue as much as the foreign companies do. These are all issues, we have a senior group of people that spends a lot of time on. I’m in China twice a year, I’ll be next week in Beijing, it will be certainly a topic I talk about. But, in a sense, it winds up almost always being more about how do we all help a Chinese software industry grow, so that there’s enough local interest in the issue, because without local interest in the issue the Chinese government has got plenty on its plate, and this doesn’t necessarily bubble up as a top priority issue.
STEVE SHEPARD: Even where they have indigenous industries, companies in other industries, piracy is more ingrained in their culture than it is in ours. They just imitate things, and it’s hard to stamp it out even when they have their own Chinese companies.
STEVE BALLMER: One of the reasons I think we’re excited that China is in the WTO, intellectual property rights have to be respected by WTO member countries over time. It’s an important issue for us.
STEVE SHEPARD: Andy Grove, whom we both admire, is speaking out a lot lately about China and particularly India, because of the threat to jobs, software engineers, for example. And he sees tremendous technological opportunities for the United States, and he thinks India in particular is going to be the main beneficiary not the United States. How do you feel about that? Are you worried about the threat to American jobs in the software industry because of India?
STEVE BALLMER: Well, I think it would be fair to be concerned. And people focus oftentimes just on the labor rate differential, but the thing that’s probably most troubling is the graduation rate of technical graduates. I mean, the United States is number three now in the world and falling behind quickly number one and number two in terms of computer science graduates.
STEVE SHEPARD: Number one and number two being?
STEVE BALLMER: China and India. And if you look at it, a higher percentage of the best and brightest, more kids in absolute graduate in computer science in those two countries, their growth rates in graduates are going up. In the United States actually we have fewer computer science graduates today than we did five years ago.
So in a sense part of it’s just about the hubbub of labor rates and blah, blah, blah. The bigger issue is what kinds of things do we need to do to encourage more American kids —
STEVE SHEPARD: Yeah, but how does that solve the problem? I mean, if we had more software engineers graduating from American universities and those jobs are going to India anyway, we just have more unemployment among software engineers.
STEVE BALLMER: Well, no, you might look, if you told me you can hire somebody in India for $20,000 a year, which would be a great, high paying job, and the —
STEVE SHEPARD: Equivalent job.
STEVE BALLMER: — sort of equivalent is $120,000 job in the U.S., that’s a very tough tradeoff, and not for me because we like to do all our R & D in Seattle, but there are some jobs for which that would be a tough tradeoff for some people. We have a small facility in India. It will grow, but we don’t anticipate it being very large so far from an R & D perspective at Microsoft.
But if you say to somebody, look, the Indian engineer is $20,000 and the engineer in the United States is $50,000 a year, which is still a very respectable job in the United States, it’s not a $120,000 job, the whole economics of that proposition start to look quite a bit different. The communication hassle, the time zone hassle, there are many things that you say the hassle of those things, are they worth the cost savings and at some big delta some people are going to say they are and a much smaller delta people are going to say they’re less valuable and that’s partly the function of how many folks are graduating with technical degrees here in the United States.
STEVE SHEPARD: One of the standard answers against the argument about concern about India is simply that, look, if we take advantage of the resources that are available in India, the terrific brain power, it’s sort of a brain game for the United States. We can do some of the work over there, we get the benefit of their ability, we can work R & D around the clock, it will spur growth rates in the United States, it will make U.S. industry more efficient, et cetera, et cetera, and in the long run we’ll be better off. Would you make that argument or —
STEVE BALLMER: I think free trade is a great thing. I think this is a free trade argument. I understand as applied math applied to economics, the second part major, I understand the argument and I think it is a correct argument. It might be painful in the short run; it’s correct in the long run.
With that said, I don’t anticipate that we will be doing most of our new development outside the United States. I expect we’ll be doing most of it inside the United States.
STEVE SHEPARD: Why?
STEVE BALLMER: I like to have people physically close together. R & D is a large number for us, but the critical issue for us isn’t whether the R & D budget is 50 percent or 30 percent bigger or smaller; we’re not there yet. The critical issue for us is do we have the right products, are we in the market at the right time, do we have the right innovation, does it have enough value, and I’ll trade off a number of things on the plus side to make sure that we have — and it’s not that Indian people can’t figure that stuff out as well, but it helps that not only to be sort of in the maelstrom of activity, it helps to be close to the biggest market in the world. You’ve got to try things out on customers. And there aren’t that many customers in India. There’s plenty of developers, but the Indian PC market is a 20th or less the size of the U.S. PC market and so you get totally different market dynamics.
STEVE SHEPARD: Okay, let’s get through some of these questions. Is the next big thing in sight?
STEVE BALLMER: Well, we think so, but we talked about our “Longhorn” product earlier and we think it is a really big thing. The big thing from an industry phenomenon standpoint, which “Longhorn” encompasses, is this notion of what we call XML in our business. I always feel a little bit like the guy in the graduate telling his soon-to-be son-in-law to go into plastics, and what the heck is XML. XML is kind of like a common language to help systems on the Internet talk to one another in a much better form. And it will be the sort of software innovation power that I think propels us for the next five, six, seven years.
STEVE SHEPARD: What is the purpose of your cash balance and how do you manage it? You have a lot of money.
STEVE BALLMER: Yeah, we have a lot of cash and I think we’ve been —
STEVE SHEPARD: What’s the number, 50 billion, 55 billion?
STEVE BALLMER: Roughly. (Laughter.)
STEVE SHEPARD: Who’s counting, yeah.
STEVE BALLMER: No, we’ve got a lot of people counting very carefully. That’s good governance, but I’m not watching on a day-to-day basis.
But I think we’ve tried to indicate a couple things. Number one, we started paying a dividend and then we increased it. We started at $0.08 a share. We’re at $0.16 a share today. So clearly we’re trying to return some of that money to our shareholders in the form of dividends. That’s number one.
Number two, I think we’ve said that until we’re through some of our additional legal risks, we think it’s imprudent to make some other major decision, buy back dividend, blah, blah, blah, whatever, with the cash, because in a sense we don’t know what business disruption — people say, “Oh, the EU never imposes fines that big, what are you thinking about.” It’s really about what level of business disruption.
And in the meantime it is our shareholders’ money. We manage it prudently, we manage it well, we occasionally make very good investments, we’ve made a few bad investments, but by and large we’re managing that portfolio responsibly on behalf of our shareholders. And assuming we’re in a position to return it to them in some form or another over time, I’m sure we will do that.
STEVE SHEPARD: What is the disruption besides the European Union possibility? And that they say the maximum fine is 10 percent of revenue, which in your case would be a fine of $3.2 billion compared to 50, 55 billion in the bank.
STEVE BALLMER: Say product design, say we don’t understand, say something happened and we for some reason wouldn’t understand what the regulatory framework is for us from a product design perspective, the issue you brought up earlier. I don’t know what the disruption is to that. I can’t model that through until we know where that goes.
And so we’re trying to, as I said, we’re working hard to try to settle the matter, but, you know, we just say, look, there’s no reason to make some major action until we’ve worked through a lot of these legal matters.
STEVE SHEPARD: Okay, big dividend after EU declares in your favor, is that it?
STEVE BALLMER: No, that’s not a headline you heard from me. (Laughter.)
STEVE SHEPARD: Well, I wanted to ask this question, too, which is about the expensing of stock options. You decided to give up on the stock options and go to restricted stock for employees. Why?
STEVE BALLMER: There are times in a company’s evolution where I think stock options are a prefect way to compensate people and at least there is a time in our evolution where in a sense it’s a funny thing. Stock option value comes largely, economists will tell you, from volatility; that is, on a stock option you can make a lot of money even if your stock never really goes up, it just bounces up and down. And the faster it bounces up and down and the bigger the bounce is the more money people can make.
Now, a long-term shareholder is making nothing but the action holder just has to make sure that they get the option at the right time, sell them at the right time, get it at the right time, sell it at the right time, blah-blah, blah-blah, blah-blah.
In a sense it’s not very well aligned with shareholder interests, at least in a company like ours, and I would say it’s kind of a crazy way to pay your people. You should want to pay your people based upon success, what does it mean to succeed. So we moved to stock awards, which are clearly more valuable if the stock is worth more and less valuable if the stock is worth less, and for our senior people we even said the number of these you get will depend upon the number of customers we have and how satisfied they are.
So in a sense I think we’ve really tied the interest of shareholders and employees better and we’ve created a framework that’s much more stable for management and employee to all be talking about the same thing in a good way.
You know, with stock options I still say, look, if you show an option holder two doors, in one door you’re guaranteed that your stock price is going to be down five percent 10 years from now but it’s 100-percent guaranteed, in the other door 99 percent case that your stock is down 90 percent and a 1-percent chance that your stock has doubled; which door does the shareholder take? Certainly door number one. Which door does the action holder take? Certainly door number two. And we wanted to eliminate some of that dissonance also in our own system.
STEVE SHEPARD: Now, Andy Grove again is on the other side of this argument. What’s your response to him?
STEVE BALLMER: OK, I didn’t argue that what we did is right for everybody. We’re not preaching, we’re not proselytizing, we’re not evangelizing; we’re merely saying that what we did we think is right for us and as soon as you decide to do what we did then expensing stock options makes sense. Otherwise you get an apples and oranges comparison of our P & L. Because certainly these stock awards we’re making must be expensed on the P & L and if you show no expense in prior years from the stock options that you were giving, it just looks like you’re cutting into profitability.
And so in order to give our shareholders an apples and apples comparison, we moved to expense stock options, not because we’re trying to evangelize and be religious about that either, we respect the fact that there’s regulatory bodies and people should make their own decision, but with the compensation decision that’s right for us, that was the accounting decision that was right for us.
And I spent lots of time, Bill spent lots of time talking to our industry cohorts about this because we did break ranks not because we think they should necessarily have to do what we did, but it was the right thing for us to do for our employees and our shareholders.
STEVE SHEPARD: What was the best, wisest advice Bill Gates ever gave you professionally and personally? He got you to work there.
STEVE BALLMER: The best advice. I say probably the best advice he ever gave me was to drop out of school. (Laughter.) And that was a huge deal. I mean, my dad didn’t go to college, as you mentioned, and the notion that I was going to drop out of business school was just loony to him, I mean, beyond insane. And if it hadn’t been kind of for Bill’s persistence and encouragement and advice, I’d probably be selling auto insurance right now.
STEVE SHEPARD: What we all tell our kids, you know, you look around and you see Bill Gates dropped out of college, Steve Jobs, Michael Dell, Larry Ellison, all these people didn’t go to college, who didn’t finish college, “Hey, dad, I don’t have to go to college either,” what do you say to your kids?
STEVE BALLMER: I think if your kid comes to you with an idea that’s compelling, as Michael Dell, Larry Ellison and Bill Gates, give ’em some money and tell ’em to drop out. (Laughter.) And in the 99.9999999 percent case tell them to get back to work, which is what I’d tell my kids.
STEVE SHEPARD: This is an interesting question. I’m not sure I understand. Does a real-time enterprise exist today?
STEVE BALLMER: Yeah. I’ll interpret the question.
STEVE SHEPARD: Please.
STEVE BALLMER: I think the questioner is asking are there really example companies where information that is necessary to run the business is really flowing and moving and getting to the people who need it in real time to make the right kinds of decisions, and what does best-in-class look like and is it actually real-time?
To which I’d say there are companies that are better and there are companies that are worse, and even companies that we think are amongst the best, we’d probably hold ourselves up to some degree because we’re supposed to be able to be a showcase. I’d still tell you I know there are plenty of opportunities even in the best of the best to be much more real time in the way the business operates.
STEVE SHEPARD: Why aren’t there more women at the top levels of Microsoft?
STEVE BALLMER: I just finished a set of reviews of all of our businesses, the people, where all we did was talk for basically a day with each of our business leaders about people issues, who, what, where, when. We talked about diversity, we talked about how we broaden diversity, we talked about our talent pool. And in a sense I’m proud because we’ve made progress over the last few years in terms of the number of women vice president, executives at Microsoft. This was a year of again progress. I think we’re up to 14 vice presidents in our ranks out of about 100, up from 10 roughly 18 months ago, so progress.
You could say, OK, that’s still a small number and I think that’s right. I think one of the key issues is if you look at our employee population, our employee population is about 65 percent people with technical background, and if you look at the graduation rates for women in technical fields, particularly in computer science, we are not disproportionately below or above the number of women who get into technical fields, and I think we more reflect the fact that we’re looking for technical people.
What can we do, what can society do to encourage more women to get in the technical field? That’s a very good and important question where a lot of ideas are being experimented with. I don’t think there’s any one breakthrough though yet we can point to.
STEVE SHEPARD: Are the percentages higher in other countries of women graduates in computer sciences or engineering?
STEVE BALLMER: That’s a question I just asked our folks to tell me about India and China. That’s the real question, what’s going on, because that’s where, as I said, so overwhelmingly a number of graduates are. The sort of knee-jerk reaction from our folks was they didn’t think it was a lot different, but I asked them to give me the data.
STEVE SHEPARD: Go find out. That would be interesting. Is your guess that it is higher in India and China?
STEVE BALLMER: My guess is it’s marginally higher, yes.
STEVE SHEPARD: OK, well, we’ve gone about an hour so I’m going to end by just saying thank you very much for being with us.
STEVE BALLMER: Thanks, real pleasure, thank you all.
(Applause.)