Hal Varian and Carl Shapiro, both economists and professors at UC Berkeley, are the authors of Information Rules: A Strategic Guide to the Network Economy. In the book, they lay out a set of strategic rules for doing business in the information and technology industries. The rules are based on what the authors call "durable economic principles." While the topic may sound familiar to readers of the business and technology press, Shapiro and Varian use sound reasoning, case studies, well-chosen historical examples, and hyperbole-free prose to bring clarity to a subject that is too often mystified and over-hyped.

Varian is the author of popular undergraduate and graduate microeconomics texts, and Shapiro served as Deputy Assistant Attorney General in the Department of Justice's Antitrust Division. Their background in economics is evident in the book’s principal theme: "You don't need a brand new economics [to explain such things as Netscape's giving away free browsers, AOL’s pricing, or Microsoft's market position]. You just need to see the really cool stuff, the material they didn't get to when you studied economics." That really cool stuff amounts to a few big ideas with serious implications for both producers and consumers of information and technology.

Tim Simcoe, a Senior Consultant with Ernst & Young’s Center for Business Innovation spoke to the authors about their survey of business strategy for the information economy.

Hal: The shortest answer I can give you is that anything that can be digitized is an information good. A computer program, an image, or a music video are all examples of information goods. Offers that include a service component can also be an information good—where you interact with a computer program or a web page for example. Perhaps you want to call that an information service rather than an information good. The basic idea is that it does not have a material component.

Carl: In terms of the underlying economics, there are economies of scale in making automobiles or washing machines, and there are economies of scale in producing information. But for information goods, the costs of production are so low that the effects of high fixed costs and low marginal costs are sharpened or accentuated. This is particularly true with the Internet and other distribution technologies available. There is also this notion that there are many different customers with varying willingness to pay for the product. These are principles that we have seen before, but with information goods they are in sharper relief. That is one of our major themes.

Hal: When we talk to managers, everyone recognizes the importance of having a product line for physical goods like washing machines or consumer appliances or automobiles. One of the points we want to make is that to sell information goods effectively, you need to think about your product line there as well. Many of the principles that apply to information goods also apply to the industrial economy. But with information goods, there are some extra twists and turns and insights that are particularly important.

Carl: Price discrimination is very common throughout our entire economy and is almost always acceptable under antitrust laws. I don’t think that price discrimination generally raises legal flags. However, the difficult issue for a business that tries to price discriminate is dealing with customers and maintaining the goodwill in those customer relationships. If one customer finds out they are paying several times what another is paying, they are unlikely to sue, but they are not likely to be very happy about it.

Hal: Just look at the airlines. You can pay $700 for a round-trip flight to England with a Saturday night stay-over and two weeks advance purchase, or you can pay $7,000 for a seat on the same plane. The marginal cost of putting someone in an empty seat on that plane is next to nothing—maybe the cost of a few bags of peanuts and pretzels. The airlines haven’t seen a lot of legal trouble for this kind of pricing, but they have to work very hard to build customer loyalty.

Carl: Lock-in is not a new phenomenon, and in many cases technology managers cannot avoid getting locked in. Boeing’s managers had to decide whose engine they were going to put on their planes and would design the plane to go with that engine, whether it was GE or Pratt & Whitney. Today, technology managers have to pick the hardware, software, databases, and information formats for the way they want to run their business. They cannot avoid lock-in. What we hope to do is help people recognize when they are getting locked in before they do so.

Once you recognize that you are facing a lock-in situation, it is also important to look at the total cost of ownership. The total costs include replacement, maintenance, supplies, follow-on purchases and services over the entire life-cycle of a product. We also make a distinction between lock-in to a technology versus lock-in to a single vendor—which is much more hazardous for the customer. Even if you are locked in to a particular standard, you still have options if you can turn to multiple vendors. Understanding the total costs of ownership, and recognizing lock-in helps customers keep their options open, or in the worst case ask for some sweeteners up front to ease the pain.

Hal: If you are lucky enough to be in a market where you have lots of competing suppliers who are supplying pretty much substitutable technology you do not have to worry too much about lock-in. The lock-in is going to occur when you may have competition up front but once you have committed to particular technology or brand, you find that you have only single or small number of suppliers.

The simplest case that we discuss in the book is the ink jet printer. These printers are selling at incredible prices, $150 to $200 for very high quality equipment that looks like a great deal. Then the first ink cartridge runs out and you find that the next one is going to cost you $60—for a $150 printer. What people do not realize is it that printers are cheap precisely because the cartridges are expensive. Manufacturers can afford to compete the margins down to almost nothing on the initial printer purchase because the margins on the "locked-in" follow-up purchases are so high.

Hal: I think we are both rather concerned about a number of the software patents that have been issued. This is something that will have to be sorted out. There is no question that many of the legal and institutional structures of the Internet are going to be in flux for a while, but the issue here is that there needs to be a workable standard. Patent life is a nice quantitative concept that you can pin down, but the standard of an invention’s novelty is a much looser concept.

Carl: I just think it is a risky area. The patent application process is ex parte, which means that there is no adversarial procedure when one gets granted. So what happens is, someone goes in and works with the patent office to establish these very broad claims. Once that person has the patents, they wield them against others who were unwilling or unable to participate in the process. The other participants’ position is usually, "Hold on ‘patent guys,’ we were already doing this!"

Hal: Of course the defendant accused of violating a patent can go to court and say that the patent was not novel. But they are behind the eight ball. Are they going to pay a royalty? Do they have the money? Could a smaller business be forced to knuckle under? It certainly happens. So there is a stifling effect when patents are granted too broadly, and it happens in part because the PTO is not very experienced in these areas.

Carl: This is a real concern for companies both big and small. I think it is critical for the PTO and the courts to sort this out.

Hal: Yes, and the situation is a classic dilemma, because everyone wants their own IP protected but not the other guy’s. So now, people are doing defensive patent filings—not because they really think an idea is worthy of protection, but because if they do not patent it, somebody else will come in there and do it.

Hal: The important point here is that a company’s goal should not be to protect their intellectual property, but to capture as much value from it as possible. One example we use is Adobe, who developed a really good page description language. Because there are some network externalities in the market for document creation—a document standard is more valuable the more readers that can view the documents—Adobe released a big chunk of their product. But they also had an implementation that was superior to the competition’s, and this allowed them to keep a few little aspects of the technology for themselves and capture the value of those components.

Carl: It is a delicate issue, how much control you maintain over a standard versus how open it is. They maintained control over the standardization of their format, but they did not maintain control over other people's implementations.

Hal: And of course, later on, they expanded into adjacent markets as a way to get more value from the intellectual property, expanding from page description into tools for creating pages like Illustrator and PhotoShop.

Carl: You have to ask yourself, what does the term "new economy" mean? At the macro level, some people say that we are never going to have another recession. That is rubbish. At the next level down, we certainly recognize that there has been a dramatic shift among sectors in the economy. Today, manufacturing only employs about 17 percent of the US workforce. That is a dramatic change from fifty or even twenty years ago.

We look at markets from the company or business unit perspective. Our book is about the difference between the market for information goods like computer software and the traditional market for manufactured items. In the market for information goods, there are issues of compatibility, high fixed and low marginal costs, and positive feedback. These things are becoming much more important because they govern sectors that are becoming a bigger part of the economy.

However, if we look back at the telephone system one hundred years ago, the advent of the radio network, or the adoption of color TV, we can see that these problems existed before and that the fundamental economics have not changed. There has been a bunch of work within the field of economics over the last twenty years to analyze these ideas, and we have contributed to that. But if you look at what certain businesses in those parts of the economy where these problems have been important were doing for the last hundred years, you can see the handwriting on the wall in terms of the strategies now being adopted.

Hal: I think there are two pieces of history. One is the history of the phenomenon, which Carl covered and that we discuss in the book. Then there is the intellectual history of economists investigating this area, which I would say the goes back about twenty-five years to where people first started to think about network affects.

Hal: Fewer and fewer actually. I have an undergraduate text book in economics, and I put a whole chapter on network affects into the last edition. That is meant for a sophomore level course.

Carl: I just taught an introductory course at the MBA level, and I gave a few lectures on network effects, lock-in, positive feedback, and switching costs. The discipline and the teaching is moving slowly towards what is more important within the underlying economy, and we see ourselves as part of the communication that is needed in order to make that shift.