Search This Blog

Sunday, March 20, 2011

Information Rules: A Strategic Guide to the Network Economy by Carl Shapiro, Hal R. Varian




Notable quotes

  • “Ignore basic economic principles at your own risk. Technology changes. Economic laws do not.”
  • “We won’t tell you that devising business strategy is like restoring an ecosystem, fighting a war, or making love. Business strategy is business strategy”
  • “If you think your position as a market leader is totally secure, try reciting the following mantra three times: ‘CP/M, Wordstar, VisiCalc’“.

Summary

This book discusses the principles behind the economics of “information goods”.

Prices

In the language of economics, information involves high sunk costs, and very low marginal costs - in other words, information is expensive to create, but costs next to nothing to reproduce, and there are virtually no limits to the amount of copies that can be produced. Information commodities trend towards the marginal cost: zero, meaning that selling information commodities is not likely to be successful in the long run. There are two sustainable structures for an information market:
  1. In the dominant firm, the largest company in the market has a cost advantage through economics of scale.
  2. A differentiated product market, like the film industry, involves a number of firms producing the same type of information, with many different varieties.
This leads to the conclusion that the strategies to follow are to either differentiate your product, or be the cost leader in your market. A price war may lead to a victory that isn’t worth winning because of the extremely slim margins. If you are the dominant player in the market, don’t be greedy - it’s often better to sacrifice some of your margins in order to keep out competitors. Don’t start a price war unless you are sure you can win, but if you are, be seen as willing to “drop the bomb” and slash prices. By sending this signal, you make other companies wary of encroaching on your territory, knowing that they won’t recoup their investments if you cut prices to the bone.
You have some room to maneuver with regards to pricing and your product if you are fortunate enough to have a unique source of information. To capture as much value as possible, personalize your product to its target as closely as possible. Targeted on-line advertising is an example of this.
Whatever your business, knowing your customer is valuable, and need their feedback in order to improve, and better target your product. The internet makes it much easier to collect current and pertinent data about your customers.
Price discrimination is the art of attempting to match as closely as possible the curve describing what people will pay for your product. A few people will be willing to pay a lot, but if your price is high, you exclude all the people willing to pay less money. On the other hand, too low a price, and you lose out on the money the high spenders would have been willing to pay. There are three ways of implementing differential pricing:
  • Personalized pricing - each user buys at their own price.
  • Group pricing, where different groups (students, retirees, etc) get their own price.
  • Versioning - different versions of the product that the user can select from.
Many web sites, especially airlines, and even amazon.com, change prices for each individual on the site based on their behavoir. Information products sold on the internet offer a great deal of flexibility in setting individual prices.
Reasons why it may make sense to target groups rather than individuals.
  • You can offer different prices to different groups depending on how sensitive they are to prices.
  • Network effects: by exploiting the fact that a product becomes more valuable the more it’s used, by offering incentives to standardize on a single product, it may be possible to acquire an entire market segment. Network effects are covered in greater detail later.
  • Lock in: Once a product has been selected and acquired, changing may be very costly.
  • Sharing: it may be easier for a group of people to pool resources to utilize information goods.

Versioning

The concept of versioning refers to the practice of creating different versions of the same product at different price points. In the world of books, hardback versions are usually released some time prior to paperbacks. Those who are willing to pay more will purchase the hardback, while those who are more price sensitive may choose to wait for the paperback. In this way, the publisher maximizes revenues, because customers self-select into groups who are willing to pay more or less.
The key to versioning is to select the aspects of your product that are important to some users, and less so to others. Several strategies that may prove effective are:
  • Delaying cheaper versions. Real time stock information costs money, whereas free quotes are available after 15+ minutes.
  • A more elaborate user interface for higher paying customers. They’re the experts, usually, and want a pleasant, efficient experience.
  • Image Resolution.
  • Speed - cheaper products are slower (sometimes artificially so, so as to be able to differentiate between purchasers).
  • Flexibility - often times more expensive products are more flexible in what they permit the user to do.
  • Capabilities, features and functions.
  • “Nagware” - it’s free to use, but pops up windows and reminders within the program itself that since you downloaded it for free, and are enjoying its use, perhaps it would be good of you to pay the registration fee.
  • Support - if you actually pay, you are entitled to some support, for free.
Customers may find the concept of versioning annoying, but remember that that segment of the market might not have been served at all were it not for a cheaper, reduced feature version.
Another useful tactic is to utilize people’s “extremeness aversion” in order to steer them away from a cheaper product. By creating three versions, people will naturally tend towards the middle one as the “safe” choice, even in cases where the low end version would meet their needs.
Bundling can be an effective strategy, where one of the included products is not something the customer would have been likely to buy by itself. An example is the Wall Street Journal on-line edition, which is offered at a low price to people who are already print subscribers. If, on the other hand, the customer would likely buy both products at their full price, bundling is likely to lose money.
Promotional pricing is a technique used for price discrimination that works because of the “inconvenience” involved. A sale might only last a few days, and you may have to wait for it to occur. Rebate coupons are often not redeemed. Through this inconvenience, people self-select - those who can pay the full price do, but there is still an opportunity to collect money from those who can’t or won’t and are willing to put up with the inconvenience to get the better price.
Be wary of getting into a competition based only on price, though, because you risk turning your product into a very low margin commodity.

Intellectual Property

Content publishers have traditionally had to content with two costs that are significantly changed by digital technology: reproduction and distribution.
Rather than fighting lower distribution costs, you need to make them work for you. By giving away free samples, you are likely to increase sales. Make sure the free samples direct customers back to you, though. Information that is time-sensitive is more naturally easier to defend against illicit copying. By the time it has been copied and illegally distributed, it may not be relevant any more.
Lower reproduction costs are not an entirely new phenomenon, either, despite the fact the digital reproductions aren’t just good, but perfect copies of the original. Once upon a time, in the 1800’s, libraries were viewed with suspicion by the publishing industry. However, the availability of books to read at a low cost drove many people to learn to read, thus expanding the market for books. Video rentals in the 1980s followed a similar pattern, with videos being expensive to purchase, and vcr’s costing hundreds of dollars, video rental stores made the technology available to more people. Initially, Hollywood was unenthusiastic about this development, but eventually reaped huge benefits.
In order to determine the ideal rights management program, an analysis of the demand curve is in order. The more liberal terms, the more copying and sharing there will be, but it will also be more valuable for your customers, moving the demand curve up. Tighter restrictions will reduce demand, but also reduce illicit use of your product.
It’s also important to take transaction costs into consideration for the “hassle” they create for customers. The lower they are (the easier it is to purchase your product), the better.

Lock-in

Lock-in is the rule, rather than the exception in information industries. Think about changing cars vs changing operating systems. Buying a new car is easy, even if you purchase a Toyota instead of a Ford. Moving to Linux from Windows is a significant challenge even for an expert. To understand the potential for lock-in, “look ahead and reason back”. Even small switching costs can have a large effect. Consider changing email addresses, for example - even though hotmail and gmail are both free, you may not wish to change from one to the other even if you like the service better, because your old email is stored with the other account, and in any case, that’s the email address that most people have for you. “Switching costs measure the extent of a customer’s lock-in to a given supplier”. Companies need to look at the total switching costs - what the customer pays, and what the company pays to acquire a new customer. For instance, with long distance carriers, simply offering a customer 10 dollars to change merely transfers the costs from the client to the company, meaning there is no net reduction. Offering free minutes, on the other hand, where the client values them more highly than the company, creates a net reduction in switching costs. If a business has high margins, this kind of offer is an attractive way of reducing switching costs. If you can easily measure the switching costs, you can calculate your expected profits from a customer as equal to the total switching costs, plus the value of the value of other competitive advantages in terms of a superior product or lower costs, with regards to rivals.
Types of lock in include: contractual commitments, durable purchases (purchasing expensive, durable equipment that can be used for many years), training specific to a particular product, data conversion costs for information and databases, specialized suppliers, search costs (finding other suppliers) and loyalty programs.
The product cycle for lock-in looks like this:
  • Brand selection - looking at what options are available.
  • Sampling - testing out a particular option.
  • Entrenchment - switching would be expensive.
  • Lock-in - switching costs are prohibitevely expensive.
As a customer, you must understand lock-in in order to deal with it effectively. The two key elements in your strategy are to strike a tough bargain at the beginning of the lock in cycle when your choices are still open, and throughout the lock-in cycle, to attempt to minimize switching costs. By bargaining initially when you have levarage, you improve your situation further down the road.
Sellers desire locked in customers, and to acquire them, need to:
  • Be prepared to invest to build the initial customer base. Smart clients will know that they face lock in, and your competitors will also want locked-in clients, so competition is fierce to offer attractive terms for the initial purchase.
  • Aim to entrench customers by creating products that your customers will have a stake in and be prepared to invest in.
  • Leverage your existing user base. For instance, influential buyers are worth even more, and should be offered corresponding discounts. Consider selling complementary products, as well as access to others to your installed base (be careful not to irritate your users by going too far, though).

Network effects

Whereas the firms of yesteryear - petrochemicals, cars, chemicals, steel, and so on - were relatively stable oligopolies where a few companies dominated an industry, temporary monopolies are what is observed in the world of high tech. The difference is in the economics of scale, versus the economics of networks and positive feedback, which reinforces the strong, and hurts the weak. This tends to create one or a few very strong winners, and few other companies in a market. The “old” companies worked with supply-side economies of scale - whoever could grow the most could cut costs and grow profits because of the efficiencies created by their massive size. Information companies instead work with demand side economies of scale: people value a product because a lot of other people use it. This creates a virtuous cycle for winners - since a lot of people use your product, even more will decide that they need to as well. Products with fewer users lose those users to products with more users, falling into a vicious cycle.
Economists use the term “network externalities” to indicate the effects of belonging to a network that creates value for its users. For example, someone buys a phone who didn’t have one before. That makes the phones of her friends and relatives more valuable because they have one more person they can call with their phones.
Of course, not every market will be dominated by these effects, leaving a lone winner. If there are widely accepted standards that allow people to switch platforms, there market will not “tip”. The markets for operating systems is dominated by Microsoft, but the market for computers is not dominated by any one company, because the owner of a Dell can interact with the owner of an IBM in terms of sharing files, email and so on, with no problems.
The two strategies that companies utilize in an attempt to deal with network externalities are an evolution strategy in which a migration path is offered, so that users of the network do not have to ‘jump’, making it easier. The ‘revolution’ strategy is painful for users in that they do have to make a leap from one platform to another, incompatible one. You must offer a strong incentive, such as vastly superior performance, in order to give them an incentive to make this happen.
Another important balancing act is in openness versus control. Make an open standard, and more companies are likely to make interoperable products, growing the market for everyone, but leaving you with a smaller portion of it. By controlling it too rigidly, you risk controlling all of a significantly smaller market. Examples abound, such as the IBM PC (open), vs the Apple Macintosh (closed).
Battles over “networks” have been fought in the past, and there is much to learn from them. Instructive case studies are to be found in the telephone industry, as well as color television.

Cooperation strategies

Competing in networked markets presents problems that are different from those in content markets, let alone “traditional” markets. Standards are almost always positive for users, because they create a larger network, reduce the uncertainty of a purchase, reduce lock-in, and create competition within, rather than for the market. Depending on the standard and what it specifies, competition is more likely to move towards price, rather than features. Competitors in a market with a standard have strong incentives to add their own, proprietary extensions that enhance the value of the product in order to make a play to extend their control in the market.
In a standards market, winners include consumers, and producers of complements to the standard product. Incumbents often view standards as a threat. Useful tactics in a standard setting context:
  • If you can move faster than your competitors, participating might not be in your interests - the standards process may only slow you down.
  • Keep moving, and don’t suspend development during the standards process.
  • Keep an eye out for potential side deals.
  • Carefully search for any patents that key players may hold that will give them an “ace in the hole”.
  • Vague promises mean nothing.
Allies can be very useful in standards battles, because a larger overall market for everyone results in a net win.

Standards wars

A standard war is what happens when two incompatible technologies clash. VHS versus Betamax is probably one of the most famous. How can you come out ahead?
Standard wars can be classified according to the relative incompatibility of the two technologies. Where one technology is incompatible, that signifies a ‘revolution’ strategy, against an ‘evolution’ strategy when backwards compatibility exists.
The capabilities and assets necessary to win a war in a network market include:
  1. Control over an installed customer base.
  2. Intellectual property rights.
  3. Ability to innovate.
  4. First-mover advantages.
  5. Manufacturing abilities, if supply side economies of scale come into play.
  6. Complements for the product in question.
  7. A strong reputation and brand name..
Whether you are pursuing an evolution or a revolution strategy, two important tactics to utilize are: 1) preemption - by being the first, you are more likely to be able to put the positive feedback cycle of network externalities to your use. 2) Managing expectations - by sending signals to the market that your product will be very popular, people may avoid purchasing a rival product.
Winners of a standards war can’t rest on their laurels, and must be wary of future developments, because the world of technology moves quickly. You also need to pay attention to your product’s complements, preferably by creating a competitive market that you don’t interfere with. Tactics to defend your position are also important, although you must keep in mind antitrust laws. Attractive terms for important complementors is a common tactic. Many companies offer discounts and assistance for their developers. Exclusivity provisions in contracts are powerful, but walk the line of what is legal. Another way to stay ahead is to develop an open standard, but by means of an intimate knowledge of the technology, develop prioprietary extensions and features that might put you in control should they become de facto standards at some point in the future.
Should you fall behind in a standards war, it is usually impossible to regain the dominant position in the market, so your best bets are either in a niche where you are strongest, or to bide your time and aim to win in the next generation. A common strategy for weaker players is to create ways of interconnecting with the dominant player (for example, emulators that let you run another operating system’s programs on your operating system). Slashing prices may be a tempting tactic, but it should be avoided because it sends a signal of weakness to the market and has historically not been effective. Lawsuits are a last resort, if the leading firm has promised openness and not delivered.
At times, a fierce battle will kill or cripple a technology altogether, so be careful not to win the battle and lose the war.

Policy implications

With some ideas about the economics of information, a discussion of government policy is in order.
Price differentiation is covered by the Robinson-Patman Act in the US, which says that differential pricing is permitted only if it doesn’t lessen competition. However, it’s clear that differential pricing is quite common. The key points to remember are: you can lower prices based on lower costs, you can set differential prices to respond to the competition, and differential pricing is only dubious should it “lessen competition”.
Competition policy is relatively vague, with the critical concept that a monopoly may arise from competition, if a firm offers lower prices and better quality - “it is not illegal to have a monopoly, only to monopolize”. Regulation may be called for only when a monopoly is unlikely to be toppled with time by new competitors.

#

Links

Professor Varian’s columns in the New York Times are good reading if you liked the book.