Andy Kessler's new book "Eat People: Unapologetic Rules for Game-Changing Entrepreneurs" is out today - and we've got an exclusive interview with Andy!
I personally learned and gained a ton from this book - as a software startup guy myself, it really helped open my mind and eyes to the patterns that lead to successful technology ventures, products, and companies. These are key lessons for anyone who works for, invests in, and or starts their own technology companies.
Like any excellent book, it triggered as many questions as answers for me - so I felt very lucky and fortunate to have the opportunity to chat with Andy and pick his brain about some of the key concepts in "Eat People."
Brett: Hi. This is Brett Owens, and we're here today with Andy Kessler. We're going to talk about his new book, which Andy was kind enough to send me an advance copy of. It's called "Eat People: And Other Unapologetic Rules for Game-Changing Entrepreneurs." I really like the book. I think it's excellent both for entrepreneurs, especially those in technology, but also outside of technology since it's the center of everything. You can really use technology to scale what you're doing in terms of your business these days.
The other thing is it's excellent for people involved in technology and people who invest in technology companies. It helps you pick out winning companies, so you can see the playbook for being successful and who's going to be basically eaten as somebody else's lunch coming up. If we can start, Andy, with your inspiration for what inspired you to write "Eat People."
Andy: Yeah. Well, you hit on part of it. I am fascinated by entrepreneurs. I've worked with tons of them over my career. Based on my experience of working on Wall Street and living out here in Silicon Valley and watching what works and watching what doesn't work, or even things that work near term and then blow up and trying to figure out why that is, so I wanted to provide a set of rules or really criteria to help entrepreneurs find the next big thing. But at the end of the day, it's not just for entrepreneurs. It's for investors, because you can find public companies that live up to the same set of rules and same set of criteria that I think gives you a leg up on finding those that are going to be successful and those that aren't. Then finally, for helping people figure out what their careers might be. You don't have to be an entrepreneur to benefit from this book. It can help you pick what companies to work for, what job, what career.
The inspiration was somewhat so I could scale myself. I was an analyst on Wall Street, then ran a hedge fund and a venture fund, have written op-eds and now books for a lot of years. I get a lot of e-mails and contacts. I try to answer as many as I can, and you can't answer all of them. The single biggest question is, "What's the next big thing?" I was an engineer-turned-finance person, so a lot of engineers and engineering students call me up and say, "Hey, I want to do what you did. What's the next big thing?" There's no right answer. I can't sit here and tell you what the next big thing is. I have some of my own ideas. But instead, I tried to write a book that laid out 12, and it's really 13 because you get a bonus rule, 12 rules of what to look for. If you can find as many of those as you can in the startup opportunity, the investment, and the job, then you're going to be highly likely to be successful. That's the inspiration for this book.
Brett: Let's get into it. A core theme in the book is scale, looking for scale. I know you've had a lot of success in your career finding things that would scale ahead of the curve. Let's get into that. What defines scale? What do you look for in companies that you invest in now or have in the past in terms of the key factors that go into scale?
Andy: To me, scale means two things and they're related. The first one is that you're not doing something that's one-off. You're not doing it for one person. You know, consultants, lawyers, or doctors, they're sort of one-off professions. You do it for one company or one patient, etc. What scale means is you do it for a million people, tens of millions of people. You write an iPhone app, and you're not writing it for some narrow market. You're writing it for millions and millions of people. Something that scales just scales as wide as possible. The way that you do that and the real criteria of scale to me is something that goes down in price and constantly goes down in price.
I was a semiconductor analyst on Wall Street. I figured out pretty quickly that semiconductor components, not so much the components, but each transistor, each bit of memory, each megahertz of performance on a microprocessor, goes down by 30% a year. The good news is that it's elastic. Every time the cost comes down, some new application opens up to take advantage of it.
It's not just in the tech world. This thing's been going on forever. Rockefeller gets beat up for busting unions and all sorts of weird things later in his career. Early on, in 1865, the price of a barrel of oil, 42 gallons, was $25. Seven years later in 1872, it was $10. By 1884, it was $3.36. One hundred and forty years ago, oil and energy scaled. It doesn't anymore. Although hopefully, solar or some of the alternative energy things will scale, but they haven't proven to me that they have yet. Rockefeller found a business that scaled and he made a fortune. By the way, he made a fortune not only for himself but for society, because for $3.36, you got rid of candles in your home or whale oil. Whales were rejoicing when Rockefeller discovered oil.
Let me give you another example with Carnegie and the Bessemer process of making steel. The biggest customer was railroads. In 1867, there were wrought iron railroads, which is like what it sounds like a wrought iron fence. It was brittle and they'd fall apart and you have to replace them all the time. Or there were the Bessemer ones which stood the test of time. The wrought iron rails were $83 a ton. Bessemer's were $170 a ton, so twice as much. Bessemer had 1% of the market in 1867. In 1884, no one made iron rails anymore. Steel rails had about three or four times the production of the wrought iron rails before at $32 a ton. So it was a 60% drop in cost over basically 17 years. Carnegie made a fortune for himself and for society who can move things and people around, etc.
Brett: Can you talk about raising versus lowering prices? You alluded to that in the book where Wall Street likes to applaud companies that raise prices or at least in the short term. That's always been a play. Anheuser-Busch (BUD), they raise beer prices and people still pay for it and it's money in their pocket. But you found examples where that really is not what you want to be doing. You really make the money when you're lowering prices but your costs are going to the floor faster.
Andy: You want to find things that are elastic. To me, scale means elasticity. If someone's raising prices, they're not elastic at all. They're inelastic. If you raise cigarette prices, eventually people will stop smoking but not really. Beer prices, I don't care what they charge for beer, I'm still drinking it. They're not going to sell twice as many by lowering the cost. If they lower their cost by 30%, they're going to have a 30% smaller business. That's the other end. You can go invest there. Warren Buffett invests there. Good luck. Have fun. I don't want anything to do with it. I want to find that riskier end where people are lowering prices, doing good for society, because we all benefit from cheaper iPhones and iPads and everything else. It also creates huge, 10x-type returns instead of 8% per year type of returns. If you're an investor and just want to throw your money in the stock market and throw your money in some old industrial or old consumer product, go ahead. You're not going to make the 25% to 40% returns that I think you could make if you select the right set of companies that scale.
Apple's (AAPL) been the classic example of scale. They've got an iPad. You couldn't have even done one three or four years ago. It would have cost 10 grand or something. You just couldn't make the damn thing. Now for $300 or $400 or whatever it is, they're changing the landscape and doing away with laptops. That is what scale is.
We do have companies that raise prices even in the tech business like eBay (EBAY). eBay discovered this wonderful way to connect buyers and sellers in a quasi-auction like thing. They had the tiger by the tail. They grew and grew and grew. Instead of continually lowering prices, they started raising prices to make up for, I don't know, revenue growth that the stock market had predicted. Eventually it broke, because if you raise prices, then someone figures out some other cheaper way of doing it to undercut you. That's a great lesson for any tech company out there. Just constantly lower prices.
There's a great quote from Jack Welch at General Electric (GE). This is such an old quote. I got it from someone that worked at GE. I don't think this was on an analyst call or anything. The famous CEO of GE would get up in front of his troops and the sales force and say, "Any asshole can raise prices. It takes guts to lower them and gain share and own the business, etc." Scale, that's it. That's number one for me.
Brett: If you're eBay and you're sitting on that cash cow that they had back in the day, the play you would have recommended for them is to continue to lower prices and to eat their own lunch but to figure out a way to make that work and to grow the overall market rather than eventually choking out their customer base and getting eaten alive by the next hungry . . .
Andy: Yeah. I think if they had lowered prices, Amazon (AMZN) wouldn't be the behemoth that it is today because they would have found their way into traditional retail markets, not just this garage sale kind of thing. They started doing it. They're doing it with autos. They should sell some meaningful percentage of autos in the world. They don't because it's too expensive to do. I just think it was a major mistake raising prices. No one quite knows that story. They go, "Oh, eBay just ran out of market." No they didn't. They raised prices and forced themselves out of new markets.
Brett: I was curious to get your take on Microsoft in that light as well. Maybe they didn't raise prices, but they kept them ... they were always pretty high. They had a monopoly, so they always got the price for Office or for Windows. Same type of story there with Microsoft (MSFT) do you think?
Andy: Absolutely. Microsoft Windows, the OEM versions were under $50. When PCs were $1,000 or even $700 to $1,000, they said, "Why are we only getting $50? We provide all the brains of this thing." They started making it up by charging hundreds of dollars for Office. But then competitors, it was Borland and then it was the company out of Utah, the other word processing company and of course Lotus with Lotus 1-2-3. They started doing these $99 upgrade prices. So in effect, it lowered the price of Office on the margin to $99. Microsoft said, "Well, we can raise our operating system price because we're not getting enough of the value in this market." That's fine. It kept them growing for a long time.
When you sell one of these netbooks for $300, when they charge $50 to $100, and I think the OEM price is less, for the operating system, that's a third of the cost of the whole thing. What happened is these netbooks got undercut by smartphones and by tablets and the iPad because Microsoft had charged too much. I think Intel (INTC) had a little bit of that problem. They had their Pentiums and they solved that by selling these Atom processors at a lower cost. But now Intel is even getting thrown out of some of that market with ARM chips. I think Intel did a two-tiered market. I think Microsoft just said, "No, we have Windows 7 and you're going to pay us $100." People start looking elsewhere for how to do their computing because they got greedy.
Brett: Yeah. Now you got them both trying to get into the mobile market with a different play. Right?
Andy: You're not selling a $300 Pentium into an iPhone.
Brett: Exactly. That's a good lead in on the next rule we want to cover here. What Microsoft did do well back in the day is they owned the operating system which is just one component of the overall system. That had to do with getting horizontal, which I think was inspired by a good quote from an old college roommate of yours as well.
Andy: My old college roommate, Franz, would put the TV on and then pass out. Right before he passed out, he said, "When in doubt, get horizontal." I would say, "I want whatever he's having."
So many companies are vertical. They do everything from soup to nuts. IBM (IBM) used to do everything. Even on their mainframes, they still pretty much do. They design the chips. They wrap plastic around them. They wrote the operating system. They wrote the applications. They have the sales force and service organization. They did everything. Innovation couldn't find a way into this vertical organization because it was slowed down by the slowest link. If it took an operating system five years, then who needed the next set of chips?
AT&T (T) was a vertical behemoth. They would sell phones. They developed the switches and the modems, handled your calls, and figured out how to price local and long distance to maximize their profits. Ford was another one. There was a Russian River plant that literally made everything. I think they had to import the rubber from Indonesia for the tires, but everything else they had there. They had the iron ore. They made the windshields and the glass. I don't know if they had cows for the leather seats.
The modern style and the one that I think you can maximize profits and have longevity is you just pick a horizontal slice. If you can find a protected horizontal slice, then you've got a huge opportunity. Intel owned the microprocessor space. They ended up with 90% of the market selling microprocessors into PCs. They didn't sell PCs. They started making boards to make it easier for PC companies. But it was Compaq, IBM, Dell (DELL), and all these other guys who would wrap the plastic around it. Of course, it was Microsoft's operating system. Increasingly, it was Microsoft's applications, but there was Lotus and Borland and some of the other companies we mentioned. Intel didn't make hard drives, so there were other companies that made hard drives. There were other guys that made just the read/write heads on the hard drives. The whole thing went horizontal.
I once came up with, and it's probably too many because it's consolidated a little bit, 30 different layers. Even in the software, you could own a layer for encryption or for doing a whole bunch of protected little software applications until Microsoft figured out how to do it and then they would squash you.
The Internet went horizontal. Instead of just AT&T handling everything, there were companies like Cisco (CSCO) that would do the routers and other companies that would provide the fiber optic cables and companies like Yahoo! (YHOO) or Google (GOOG) doing the application layer above that. So you didn't have to do the whole thing.
Apple looks somewhat like a vertical company and they are because they have stores and they do wrap the plastic around it, but a lot of the stuff they do outside. They moved the manufacturing outside and the chip making others do. There is some verticality to what Apple does. They like to protect things. Some day, it's going to bite them in the butt. It just will. But in the meantime, the value added horizontal, which is the iOS operating system, at least the baseline applications, the form factor, and the user experience is the horizontal layer they own. I think you could find these horizontal layers in tons of places. They're all over the technology marketplace.
When I look at investment opportunities, private and public, I go, "Is this a horizontal or is this company trying to do the whole thing from soup to nuts?" If it's soup to nuts, it may work near term and it's going to blow up some day. If it's horizontal, you better find the one that owns that horizontal and it's protected in some form, and they have the ability to turn out faster and higher performance versions of it, etc. But being horizontal, I think, is absolutely the way to go.
Brett: It seems like that's increasingly the way to go as specialization takes over. I've noticed that in software where you've got small, sort of lean companies that can attack one really, really narrow problem and do it really well. The, like you said, with the Internet, they can open their API and make it possible to import/export data pretty seamlessly out of there. You head off the big Microsofts where their big dev teams are more of a disadvantage now than the advantage.
One story I did want to cover off on, on the horizontal that you mentioned in the book. You sat next to a guy on a plane doing the SPF compounds. Perfect horizontal play in your mind and he ended up trying to take that thing vertical or said he was going to take it vertical.
Andy: I still haven't seen it in the marketplace. In the book I tell a story of, as I'm writing this whole chapter on getting horizontal and all the great examples, I am sitting next to a guy on a plane who starts telling me about his company on the East Coast somewhere that has this compound that can always have, I think, a negative charge. What they figured out how to do is apply it to sunscreen. They can use this compound and it'll be negatively charged, so they can have it in your soap or your shampoo. Think about your soap. You take a shower in the morning, you use the soap, and for the next day, because of this negative charge on these SPF compounds, it sticks to your skin. So instead of the oily crap in suntan lotion that we slop on and we feel slimy all day, this is just sitting there with a charge. Sure enough, you have sunscreen protection, which is a great way to hold off skin cancer and all these great things.
I said, "Oh, that's great. I can imagine Ivory soap with their compound inside and Irish Spring and whatever all the other soaps are." The guy goes, "No, no, no. We're building a factory. We're going to build our own soap." I just sat there and I go, "You're crazy because you're going to have to fight for shelf space against companies that have been selling soap for almost 100 years. They own the retail space and they own the branding. You're going to have to spend hundreds of millions to get your branding when you can just leverage off of someone else's brand." This was a couple years ago. I don't know what happened to that company. On the other hand, I went to a store and looked for a sunscreen soap. They don't exist.
I think if he had gone horizontal, it would be tough work to weasel your way in to some of these bigger companies. But if you can become a sub-brand underneath a big one because you do one thing really well and are horizontal, I think it's a much higher chance of success.
Brett: Sure and let them do the marketing, manufacturing, and the sales end for you.
Andy: Yeah. I'll give you one other example, and we can move on to another one. Google Android. This ties into another rule that I call Zero Marginal Cost, which I'll just touch on briefly. They developed a viable operating system for smartphones. At first, they made the Google Nexus One or whatever it was. I think it was just a sample phone. They sold it off of their website. Instead what they ended up doing was just giving it away. You want to do a smartphone? Here's the operating system. It doesn't cost us anything. It costs on a marginal basis. We have all the engineers that develop it and keep it up. If you sell one copy or 10 million copies, because we gave it to you for free, it doesn't cost us any more.
Of course the way that Google makes up for that is they're giving something away at a horizontal layer below them either upstream or downstream. You have to figure that out. They're going to make money by having that operating system direct more search queries to the horizontal layer on the Internet that they do own, which is search. Sometimes you can give something away if you own one of the horizontal layers or slivers upstream or downstream from what you give away. I think that's a perfect example and a great example. Some day, that's going to go and bite Apple. We talked about something is going to bite Apple in the butt later on. That's one thing that I would say there's a probability of that happening, because someone is devaluing your horizontal layer to increase the value of theirs.
Brett: It's an interesting point. That's one people have thought Google was crazy for, for a while because they make all this cool stuff and then they just give it away. There's a method to the madness, which is to drive everybody to the search stream, which you say, which is really the cash cow, the heart of the business. As long as they're getting you in the direction, they're happy.
Brett: With that, we can't get you out without getting to everyone's favorite, the "Eat People." I've had your book laying on my coffee table with the fork sticking out of the person with a briefcase flying. Anyone who's come over has shrieked a little bit at the book. I try to explain this is just the way the world works. It's why we keep progressing. Can you get into that a little bit, the eat people, the productivity angle of things and why it's actually a good thing to knock out jobs?
Andy: It's an age old phenomenon. Technology comes along and it displaces older, lower margin jobs with better, higher paying, and higher margin jobs. It was 80% of Americans used to be farmers. Now it's 3%. Many of them moved into manufacturing. Manufacturing is partly technological and partly globalization and cheaper workforce in China, etc. Now it's a service economy. For the last 15 years, the service economy has been under attack by people eating technology. Think of tellers, operators, travel agents, stock brokers, and stock traders on the floor of the New York Stock Exchange. There's a lot less of all of those jobs than there were, heck, even five years ago, because you can use technology and the Web to displace that worker. This technology came along and it ate all those jobs and all those people. Bad news if you had that job, but good news for society because it's cheaper to book tickets, get money out of the bank, and cheaper and faster to trade stocks. There are all sorts of benefits from it.
What my view is, as we're sitting here in 2011 already, I don't know how that snuck up on us, there's a whole other class of service workers that are going to be attacked by technology. What I did is I broke the world into different types of workers. I did it initially by saying, "Look, there are really just two types. There are creators, those that are creating the tools, the productivity tools, and use those tools. Then there are servers. Servers do nothing but service the creators."
Brett: What do you have that break down? You talk about that a little later in the book, but in terms of percentage creators versus servers on planet earth.
Andy: It's probably 1 to 99. You may be a creator or you may have a service job working for one of the creators and benefiting from the profitability and equity values that are created. It could probably be more like 25 to 75. There are certain jobs we're not going to get rid of -- cooks and someone that shines your shoes. There are a lot of service jobs that are never going to be gotten rid of. But I think there are higher paying service jobs that are going to start being eliminated, and it's going to scare the crap out of people.
What I did is I divided servers, who serve creators and of course they serve other servers, into a bunch of different types. I called them sloppers, sponges, super sloppers, slimers, and thieves. I did it because I had fun with it. Each one of them you can find jobs that are going to disappear.
Sloppers are really easy. It's someone that moves stuff from one side of a room or one side of a warehouse to another, or one side of the economy to the other. Think of someone at the DMV. They move information from one side of a counter to the other. All you have to do is put their screen in front of you and you can do all the work and you could eliminate three-quarters of the jobs at the DMV. You could probably get rid of three-quarters of government jobs. There are a lot of sloppers that are going to be out of work. That's not just low-end workers. It's sales and marketing people too. Look at AdWords and look at the self-service model. Look at how Groupon is ripping through the retail and coupon business. Those that just forever have moved data or moved money from one side of a transaction to another, their jobs are suspect. I wouldn't want to be a salesperson for something that really can be sold with a self-service website.
You have to read the book to read about all of these. The sponges are my favorite. A sponge is someone where you have to take a test to get that job. You have to pass a bar exam if you're a lawyer or doctor. You have to pass a test to become a realtor or a plumber or to cut someone's hair for God's sake. All they're doing is limiting supply and therefore increasing the compensation of those that do pass the test. Increasingly, a lot of those jobs, I think, are under fire. My favorite is this whole e-discovery movement in the legal profession. Companies get sued. They're asked for 10 million documents. There's some paralegal. Someone's charging $100 an hour to partners who are charging $500 an hour to read through these million plus documents. With e-discovery, it's scanned in and you look for keywords. One or two people can sit there and sift through all of the documents in a discovery process. Law firms are up in arms, but corporations are saying, "Why am I paying you to read all this garbage? We can replace you with machines." Even these higher end jobs are under attack.
Super sloppers are someone that marks up goods more than they're worth, like Rolex. I have a watch on my iPhone they give me for free. It's a clock. It keeps more accurate time than a Rolex. You don't have to wind it. It tells me the date. It doesn't look good on my arm. I wouldn't want to wear my iPhone on my arm, but I don't have to pay these super sloppers their tax on my vanity, if you will.
It's part of the filter. If you can find companies who are getting rid of other jobs, sometimes it's not so obvious and sometimes it is. Google got rid of a lot of librarians. Microsoft Word got rid of a lot of typists and secretaries. What about Facebook? Facebook is a company that scales. It lowers the cost of communications between people. There always was someone who was doing that social coordinating. At the end of the day, even that is a people-eating enterprise.
Brett: Let's wrap it up with your old colleagues. I also enjoyed in "Eat People" the classification of the slimer group.
Andy: Slimers, yeah. I worked on Wall Street for a lot of years, and so I have a tiny bit of sympathy for what they do. Wall Street is really about providing access to capital for growth companies and more expensive capital for companies that aren't growing so fast. There's this massive stock market and bond market that corporations and individuals can't access individually. They just can't do it. I liken these markets to a giant ball of plasma. It's two zillion degrees and you can't get too close to touch it or you get burned. Wall Street has figured out how to put a wrapper around that giant plasma and charge fees for others to get access to the growth capital they need. I liken Wall Street to the grease on the gears of the economy, like Charlie Chaplin in that movie, the famous picture of him with the wrench on the gears of society, except we don't have an industrial society, we have a service economy. Wall Street provides that grease.
They're slimers because it can be a little bit of a slimy business. We don't need hundreds of thousands of people trying to shave a nickel off of every transaction for themselves in Wall Street. I think Wall Street is going to be half as large as it was at the peak a couple of years ago, but it still plays an important function in our economy. You're not going to get rid of all of them. They're still going to be highly paid. The ones that do provide the right service to help growth of companies, whoever's going to take Facebook or Groupon public, they may be overpaid for doing it but it's going to help provide them the capital to go and make those wrenching changes and eat all those people and change all those job descriptions that are going to take place over the next 20 years.
Brett: Thanks, Andy. This has been fun going through the "Eat People" neighborhood with you. We've got "Eat People" which releases today. Like I said, for entrepreneurs and business owners and investors, I would absolutely think it's a must read, because there's a lot of content in there. I think we could have gone on for about two hours.
Disclosure: I am long INTC.