A newspaper is a bundled product. A bundled product is one that combines a number of products the demands for which may be quite different--some consumers may want some of the products in the bundle, other consumers may want other products in the bundle. (Another good example is the Windows operating system, a bundle of a number of different programs.) Bundling is efficient if the cost to the consumer of the bundled products that he doesn't want is less than the cost saving from bundling. A particular newspaper reader might want just the sports section and the classified ads, but if for example delivery costs are high, the price of separate sports and classified-ad "newspapers" might exceed that of a newspaper that contained both those and other sections as well, even though this reader was not interested in the other sections. Bundling also facilitates price discrimination by snagging consumers who place a high value on particular products in the bundle. It also increases the risk of entry by single-product competitors because the marginal cost to the consumer of the bundle of any component of it is zero. He gets the sports section for "free" (in the sense that the newspaper costs him no less if he throws the sports section away without reading it) but would have to pay a positive price for a free-standing sports newspaper. Like other intellectual products, a newspaper has high fixed costs (the newsroom, etc.) but low marginal costs (the cost of printing and selling one more copy), and so there is a tendency to natural monopoly in local newspaper markets. It is offset, however, to an extent, by differences in content, outlook, and so forth among different newspapers, which limits substitutability and therefore makes some degree of competition viable. Nevertheless newspapers tend to be quite profitable (as recently as 2006, the average ratio of profit to revenue was 17 percent, which is high relative to industry as a whole), because competition is limited. High newspaper profits sometimes are attributed to the fact that most information comes free from public sources and that newspapers deal directly with their customers and so economize on distribution costs. But low costs are not a reason for high profits, since competition tends to push revenues down to costs. High profits may seem inconsistent with declining revenues, but are not if the firm, seeing no future for itself, ceases investing in the its future and instead cuts costs to the bone (thus treating the firm's product or service as a "cash cow"). Many newspapers are doing that. Still, newspaper profits are plummeting, and with them the value of the companies. The reason is declining ad revenues (an inflation-adjusted decline of 20 percent between 2000 and 2007, and a further decline this year). This is a function in part of declining newspaper circulation but more profoundly of unbundling, as unbundling is the cause both of the declining ad revenues and of declining circulation. The Web provides a virtually costless method of distributing the products that are bundled in a newspaper. The distribution is not only cheaper, but better, because it avoids the time and space constraints of hard copy delivered on a daily (rather than instantaneous) basis and space-constrained by the cost of paper. The unbundling goes deeper than the section level (classified ads, the sports section, etc.), for every section of a newspaper is itself a bundle. The news section bundles a variety of news stories that different readers value differently; readers who have no interest in foreign policy nevertheless pay for a newspaper that may maintain costly foreign bureaus in order to produce good stories on foreign policy. The Web provides a customized news service that enables the tastes of particular readers to be identified and then satisfied by instantaneous and often costless delivery of a product laser-focused on those tastes. The bother associated with the physical bulk of the newspaper is also eliminated. A study by comScore, Inc. in March of this year found that persons 65 and older are almost six times as likely to read a newspaper six days a week than persons aged 25 to 34 (and almost ten times as likely as those aged 18 to 24). The principal reason for the difference is not I think that older people have more leisure, because people in the 45 to 54 year old bracket, who do not have more leisure than the young, are more than twice as likely to read a newspaper six days a week than the young cohort. The reason, rather, is that younger people are much more comfortable getting information online than older people are; they have grown up in the electronic revolution. This will not change as they get older. It appears that the only hope for the newspapers is to go online, and they have done this and have attracted many viewers to their Web sites. But they have not been able to charge for online ads anything like what they can charge for ads in their hard-copy editions. The reason I think is that there is much more competition in online advertising than in print advertising, especially for advertising, such as classified advertising, that is primarily informational; for the information in the ads is often available online at no or nominal cost from other sources, such as Craigslist. Moreover, the online newspaper is still a bundled product, and the Web provides close substitutes for all the sticks in the bundle. The blogs are a big factor here; in the aggregate, they not only are nimbler, but contain a vastly greater body of specialized knowledge, than the newspapers or other conventional media (as Dan Rather learned to his sorrow). Suppose, then, that the newspapers are doomed, or, more realistically, that they are likely to continue to shrink, eventually becoming a retirement service, like Elderhostel. Are there social consequences that should trouble us? A common argument is that if news is customized to the tastes and interests of every individual in the society, people will not be exposed to conflicting views and as a result will become incapable of active civic engagement, for example as voters. That is implausible. It is important to distinguish between opinion and fact. Most people do not want their opinions challenged. So if they are liberal they read the New York Times and if they are conservative they read the Wall Street Journal. But people are both interested in, and influenced by, facts, such as the fall of communism or the rise in gasoline prices, and they will learn these facts (and more quickly) on the Web even if they do not read newspapers. The few people who actually read, compare, and take seriously opposing views on matters of public policy will continue to do so after they stop subscribing to print newspapers. With the rise of the blogs, moreover, the amount of information and opinion reaching the public is far greater than in the heyday of the print newspapers. A second concern, to which the rise of the blogs may be only a partial answer, is that Internet news services (such as Google News) are parasitic on the print newspapers' large staffs of reporters, so that if they drive the newspapers out of business the Internet news services will lose much of their content. The copyright law cannot prevent this, because a newspaper can prevent the copying only of its articles--that is, of the verbal form in which the information in an article is expressed--and not the information itself. And it cannot prevent a news service from simply sending the viewer to the newspaper article via a Web link. The concern, in short, is that the Internet will kill the goose that lays the golden egg. But this is unlikely. If online viewers want the level of news and opinion that print reporters generate, the Internet news services will hire reporters, defraying the cost out of their online advertising revenues, which will be greater for an Internet news service that attracts additional viewers by offering them richer, newspaper-type fare. Indeed, long after newspapers like the New York Times and the Washington Post have ceased print publication, their Web sites may be among the leading Internet news services. The aggregate amount of news and opinion may be less, however, because unbundling will eliminate internal subsidies, for example of the news and op-ed pages by revenues from classified ads.