Scott Worden

One of the most remarkable aspects of the World Wide Web is the extent to which valuable information – information that is sold for profit outside the cyberworld – is available for free. A monthly New York Times ¹ subscription costs $32, while anyone with Internet access can read the entire paper daily by simply providing a few demographic details. The Weather Channel ² , which cable companies charge to watch on TV, maintains an exhaustive site allowing net users to search for current forecasts and articles entirely free of charge. Publicity and paid advertising for these and similar sites in part justify their existence. But as the novelty of the Internet wears off, companies are likely to look to their bottom line to justify a significant presence on the Web. And with that in mind, electronic payment systems that enable companies to efficiently capture the value of information services will likely have a significant impact on the structure and content of the Web.

One of the most promising methods of collecting revenue for electronically transmitted commodities is the use of micropayments: the ability to pay digitally for bits of data or services in small increments – as little as 1/10 of a cent – that would be otherwise impracticable in the conventional marketplace. Micropayments are particularly attractive over the Internet, where electronic commerce in general is gaining a prominent foothold and where data and services dominate the market. Companies like British Telecom and Digital Equipment Corporation ³ are investing millions in micropayment technology trials, while many cyberluminaries extol the virtues of a micropayment Internet environment. Nicholas Negroponte, head of MIT’s Media Lab, has said that microcash transactions "will change consumer behavior enormously." Others envision an environment where checks and credit cards are obsolete, replaced by smart cards holding digital value in infinitely divisible increments. ⁴

For all its promise, however, micropayments have yet to make an impression on the booming Internet commerce industry. DigiCash, a pioneer in the micropayment field, filed for bankruptcy in November 1998 after several years of trying to convert businesses to the idea of charging in tiny increments. None of the other micropayment systems being tested, including BT Array, Millicent, and CyberCash, have made money on their pilot programs. ⁵ And the amount of business conducted with micropayments is negligible compared to that done with credit cards or between. ⁶

This leaves businesses and Internet scholars with the question of whether micropayments will capture a significant enough slice of the Internet marketplace to merit research time and study dollars. The short answer is yes. As a number of technological forces converge – from Web TV and cheap network computers, to smart cards holding digital cash, to trusted systems that protect digital intellectual property – a demand will arise for an easy way to pay pennies for small products and services. But it will probably take shape not as an imitation of the current credit card environment writ small. Instead micropayments will serve as an alternative system of capturing value for digital property that exists alongside traditional modes of payment like monthly memberships, advertising, and credit card bills. The result will be a necessary balancing of cyber interests such that expensive and valuable services like the New York Times or financial data will be available only for sale. Whereas sites that are interesting but have little commercial value, like John Doe’s catalogued collection of Star Trek quotes, will still be available for free.

These changes are likely several years down the road, but the following study examines how current trends in electronic commerce impact micropayment’s viability and, in turn, the potential impact micropayments will have on internet transactions. By looking at theories of Internet commerce, new web technologies, and current examples of micropayment marketing strategies, one can see that micropayments will likely find a niche that incentives small content producers to market innovative new ideas.

Three types of sites currently dominate the Web: personal pages like John Doe’s, informational pages that promote an organization’s, government’s, or company’s activities, and commercial sites operating in hope of turning a profit. The wealth of free information and the ease with which people can access it is certainly a principal appeal of the Internet and an aspect that distinguishes it from the real world marketplace. But the Internet is still in a highly experimental stage, with much of its commercial value being dispensed for free in hopes of gaining publicity for later profits. Slate, an online magazine run by Microsoft, offered its content free for over a year to attract readers before instituting a $20 per month subscription fee. ⁷ Merrill Lynch is offering free equity research on its site in an effort to draw in paying customers. ⁸ And pornographic sites employ a variety of free content to attract viewers to membership services.

Greater emphasis on revenue return is unlikely to eliminate the appeal of personal pages or purely informational sites, just as the present marketplace retains room for nonprofit advocacy groups, public service organizations, and community theaters. But businesses that drive the expanding scale of the Web and hence the appeal of being on it for personal or informational purposes, will have to find efficient ways of making money for the benefits of the Internet’s unique freedoms to reach a majority audience.

Commercial sites generally earn revenue by selling advertising, memberships, demographic information, or payment for individual services. ⁹ Advertising and membership will probably remain the most prevalent income producers for Web sites. However, with commercial volume growing exponentially – analysts have pegged 1996 consumer sales over the Internet at $500 million while Forrester research predicts 1998 Web sales will reach $7.8 billion, rising to $108 billion in 2003 ¹⁰ – the potential for micropayments to seize a sizeable chunk of online commerce is great. And regardless of the difficulty micropayment schemes have had in securing a foothold in the online commerce market it seems inevitable that technology and volume will make such sales inevitable in the next few years.

Micropayments are especially well suited for buying digital products that can be downloaded directly from the Web, including data, music, services, and software. ¹¹ Since the marginal cost of producing and distributing these products over the Web is negligible, companies can afford to sell them at very low cost, assuming there is sufficient volume to justify the initial expense. Optimists suggest that once micropayment technology is perfected, consumers will willingly pay five cents for a newspaper article, a dollar for a song, two-fifty per stock trade, or ten bucks to rent Microsoft Word until the term paper is done.

Pessimists note that individual consumers "dislike paying for information and positively hate meters," citing pay-per-view television’s middling performance and the overwhelming preference for monthly access fees to the Internet as evidence. ¹² One study conducted by AT&T found that customers preferred a flat monthly rate for local calls over a pay-per-call system even though many made few local calls at all. ¹³ If the same phenomenon holds for the Web, micropayments will lose out to memberships and advertising. Another flaw with micropayments is that frequent customers are penalized for their patronage when they pay to play. This can be eliminated by offering a variety of payment schemes, but doing so would lessen any unique effect micropayments might have on the structure and content of the Internet.

Whether the optimists or pessimists are proved correct depends on the nature of how the Web affects traditional market models. Is the Internet simply different in the quantity and speed of information that can be vended or does advanced technology represent a difference in the kind of market consumers face?

Wired Magazine executive editor Kevin Kelly argues that companies hoping to make money on the Internet should "follow the free" and emulate Netscape’s strategy of building a significant Web presence by giving information away. ¹⁴ Kelly also advocates "feed[ing] the Web first" by establishing business models that extend the network rather than the company’s share of it. He assumes that information, communication, and network relationships are the key to future business success. Therefore strategies that will maximize exposure (like a site that provides free content in return for some basic user information) and create consumer relationships are better than ones that charge for individual usage.

Two basic economic principles underlie Kelly’s conclusions. First, network effects will dominate the Web. Unlike the traditional economic production model of diminishing returns, network effects describe a situation in which the value of a product increases as more units are produced. The first fax machine was worth almost nothing, but once every office had one it became a business necessity. Likewise, web technologies will become more valuable the more people use them. ¹⁵ Second, Kelly notes that scarcity is what creates value, and now that digital information is ubiquitous and easily copied, networks and relationships are the only scarce commodities in the information economy. Once a free product like Adobe Postscript for laser printing or the Netscape browser become commonplace a company will be well positioned to market its name to companies and users behind the Internet. Merrill Lynch’s Website is another example of this phenomenon. ¹⁶

Going further than Kelly, Michael Goldhaber argues that "the currency of the New Economy won’t be money, but attention," and therefore businesses should focus on eyes rather than dollars if they want to succeed on the Web. ¹⁷ Since attention is limited to 24 hours per person, per day, it is the ultimate scarce resource that should cultivate. Attention is also a zero-sum commodity. If one company has a consumer’s attention then no one else can at the same time. Therefore "attention transactions" – instances when a producer and consumer are interacting – will become the currency of the Web. Eventually, Goldhaber maintains, communities of attention will build up in which goods and services are bartered for returns of someone’s focused time. The more attention a person attracts, the more goods she can receive in return. In practical terms, this would mean that advertising and free distribution are the key to business success on the Web rather than micro or any other type of payment.

In a contrary and more realistic vein, Carl Shapiro and Hal Varian argue that the Internet closely resembles other industries (like airlines) with high fixed and low marginal costs. Therefore aggressive marketing and clever price discrimination schemes are the key to profiting on the Web. ¹⁸ According to Shapiro and Varian the Internet presents a difference in scale, not difference in kind when it comes to market economics. Companies will have to devise ways of creating differential value in like products despite the negligible marginal cost of distributing information. Airlines do this by charging more for the same seat when it is bought just before a flight. Likewise, IBM employed a rather Machiavellian price discrimination strategy when it deliberately slowed the speed of its F series business printer to make the lower priced E series. And Microsoft’s Windows NT Workstation is identical to the more expensive NT server, save a few code changes that give the former fewer capabilities. ¹⁹

Shapiro and Varian also note the importance of network effects, but see the establishment of a dominant network as the license necessary to begin employing the kinds of price discrimination schemes that will enable profit over the Internet. Firms should concentrate on versioning to capitalize on the networked information economy. Like IBM and Microsoft, information providers should offer an array of personalized products at different prices to appeal to different consumers. They should also try to lock consumers in by offering incentives for repeat players and extracting high margins from them down the road, as frequent flyer programs are designed to do. Finally, companies should build networks by bundling products like software suites, which encourage consumers to stick with a particular brand. ²⁰

Tying each of these threads together is the question of what quantities digital information should be packaged in. Economists refer to the practice of grouping items together for a single sale as bundling. Traditionally it has been held that consumers are likely only to purchase items when the quantity reaches a critical mass to justify the effort of the purchase. So CDs usually have ten songs, magazines have many articles, and batteries come in packs of four. This arrangement has also benefited producers because the marginal cost of production generally decreases as quantity increases, and distribution costs can be greatly reduced by economies of scale. ²¹

The Internet changes the logic of bundling for digital products, however, since the marginal cost of production is reduced to almost zero and distribution costs are negligible. Therefore once a company takes the time and money to produce a digital product it can afford to offer it to the public at very low prices when sales volume is high. The question then becomes whether consumers will be willing to abandon their traditional calculations of value – that a newspaper is only worth buying if it has hundreds of articles, and then for 50 cents – for the new thinking that if you only read 12 articles it is worth paying 1 cent each for the ones you actually read. Micropayments make this second sales model possible. But conventional bundling theory militates against it.

Some scholars have suggested that bundling theory should be altered for dealing with the cyber world. One group of authors suggests that large profits can be made by grouping together huge quantities of disparate information and selling access at a relatively low cost. ²² This is essentially the Lexis/Nexis business model, where subscribers may access an enormous database of news articles and case law for cents per item or dollars per month.

Other research suggests that profits can also be made by allowing easily copied information to be circulated among small groups for free. ²³ According to this theory bundling may occur among consumers as well as goods. Small groups that know they will share a product will be more willing to pay a higher price than the total revenue generated if a few chose to buy an item at a given price while others passed on the opportunity. If so, then Kelly and Goldhaber are correct and micropayments would only restrict the valuable returns of network effects.

Several threads stand out from these three analyses that bear directly on the impact micropayments will have on the Web. One is that network effects are crucial to successful Web ventures. Therefore micropayments are unlikely to take hold unless some big players like Microsoft become involved and attract a large number of users to form a nexus of support. Another is the issue of scarcity. A great tension exists on the Web between notoriety, necessary to achieve network effects, and limited access that allows businesses to charge for services. Once a company has attracted customers it needs to prevent them from disseminating digital copies if the market is to be sustained. Trusted systems, or codes that prevent digital information from being copied, are essential for any kind of micropayment system to become viable. Finally, the issue of bundling has enormous implications for micropayments. If retailers choose to package data in suites or offer only membership service, then macropayments will suffice to pay the bill. But if consumers are willing to pay for individual digital works like articles or songs, then micropayments will be central to shaping this new market.

Two recent Web initiatives offer clues on how these issues will develop over the next few years. Smart cards that hold electronic value as a substitute for cash are on the way and could provide a gateway to wider acceptance of micropayment methods. Developing trusted systems technology will enable content providers to limit access to valuable intellectual property. Together these developments represent a paradigm shift that will could make micropayments practical.

Smart Cards:

The main problem with micropayment ventures to this point is that customers who take the time to set up an account find their purchasing options severely limited. Most micropayment vendors have to work out payment arrangements with vendors in advance, which introduces high barriers to participation. Furthermore, the process itself is technically complicated. Digital’s Millicent, for example, requires users to set up an account, maintain a virtual "wallet" of electronic scrip paid for by credit card or check, and then trade vendor scrip for broker scrip to clear their account. ²⁴ Other systems rely on consumers setting up accounts with brokers who aggregate micropayments to place on a monthly credit statement. ²⁵ But again the barriers to entry are high and the vendor base is limited.

What is needed is a way to make micropayments user friendly and ubiquitous. Enter Microsoft, which has recently launched a smart card initiative to compete with Visa and MasterCard to control the standards for issuing digital money. Smart cards work like ATM or debit cards in that they enable users to pay electronically for goods on the spot. ²⁶ But smart cards go beyond ATMs by placing a user’s entire bank account on a chip in the card. Electronic value is held by the user, not a bank, and purchases can be made without any financial intermediary. Purchases can also be made in tiny increments, which makes it an ideal vehicle for micropayments. If Microsoft can successfully link its smart card business to its increasingly popular Windows suite of programs, then micropayments may be able to reach a critical enough mass for network effects to begin. Once smart cards are established as a common mode of payment, micropayment technology will succeed or fail based on business strategy rather than barriers to entry.

Once the technology is in place to enable easy micropayments, content providers will need ways to protect digital works from being copied. Presently there is little reason, copyright laws aside, to buy a digital product when you can copy it for free. Vendors are therefore reluctant to sell proprietary information over the Web until safeguards can be put in place to protect that information’s scarcity.

Trusted systems are one way to ensure that digitized intellectual property can be restricted to paying customers. Trusted systems work by embedding a code in software products that can be recognized by specially equipped hardware instructing the computer or printer to allow only certain operations. ²⁷ These instructions, called usage rights, could specify whether a user could make copies, "loan" a document to another computer for a specified period of time, or allow read-only capability. Vendors using trusted systems then could have control over how much access to allow consumers and instill value-creating scarcity on the digital information market.

Consumers could be induced to acquire trusted system technology with initial offers of free information in exchange for installing trusted systems software. This has already occurred to some extent with the pdf format, which makes documents available in a digitally unalterable form to anyone who downloads Adobe Acrobat software. ²⁸ With pdf, users may print documents but not edit them. Trusted systems would have broader applications, including software, music, video, and interactive documents. Once trusted systems become prevalent, digital information will have value because it is once again scarce. Companies and individuals will then have an enormous incentive to produce original and value laden products because they know their work will be protected. Micropayments will be an ideal way to charge for that value.

Two examples of disaggregated goods finding demand on the Internet demonstrate the beginnings of micropayment success and the influence it may have on Web markets. Music and news archives are currently individually distributed on the Web with some success. Meanwhile advertising, which marks the biggest obstacle to micropayment success, may be moving toward incorporating microcommercial mechanisms to attract more viewers

The most interesting example of desegregating traditional bundles and trading them over the Internet lies in MP3 technology that enables compressed music and video files to be downloaded from the Web. Copyright pirates on hundreds of not-so-secret servers have established clearinghouses for downloadable digital tracks of almost any CD imaginable. ²⁹ By downloading shareware from the Internet, the savvy user can access hundreds of copyrighted songs and software for virtually nothing. MP3 files are roughly one sixteenth the size of a CD file and can be stored on a hard drive or burned into a blank CD with the right equipment.

Although there is no fee for this service, many pirating sites require that visitors upload at least one thing of value in exchange for access to the pirate database. So for the price of sharing your latest version of Eudora you can receive perfect digital copies of the latest U2 single along with programs and games. The beauty of the system, beyond its cost, is that since you do not pay, there is no incentive to take unwanted tracks. That way you can compile a greatest hits CD from among the top current albums or pick from among the Microsoft Office Suite programs that you will actually use. As MP3.com president Michael Robertson puts it, "it’s ridiculous to spend $16 on a CD if all you want is one song." ³⁰

The music industry is understandably concerned with this pirating and efforts are underway to stifle MP3 distribution sites. ³¹ Geffen Records has issued hundreds of legal threats to hosts of MP3 sites, including MP3.com, ordering them to desist in their illegal copying. ³² But the next step may be to imitate the pirate sites. With MP3 players emerging on the market that can play rerecordable, downloaded music much like a walkman, a niche market in Web music may explode, forcing major music labels with the to sell music by the track at a significant discount. A report by Forrester notes that MP3.com had 90,000 unique visitors a day downloading 50,000 tracks from 75 bands. ³³ And it predicts that the downloadable music market will capture five percent of music retail revenue in the next five years. ³⁴ If music companies can successfully introduce trusted systems in songs, reducing the price of music to under a dollar per song could open up a massive new market with significant marginal returns.

That this behavior goes on in the gray market where transaction costs are negligible indicates consumers’ desire to obtain desegregated bits of electronic goods when the cost is right. It seems natural then that as companies develop technology that would disable a pirated program, artistic work, or article, there is some point in between free and the price of a magazine or CD that consumers would be willing to pay for individual pieces of information. That price is probably very close to zero and the key is to facilitate ways to make such small payments without much effort. Micropayments would make those individual transactions possible and thus attractive to the consumer.

Another arena ripe for micropayments is in downloadable news archives. While many online papers offer access to their archives for free, searches are often unreliable and are not user friendly. Many that do have user friendly archives charge for access and you have to set up an account with your credit card in advance. ³⁵ But some companies are experimenting with pay as you go options. The Financial Times recently introduced a pay-per-view archive that offers searchable access to four million articles online in addition to its free daily content. ³⁶ The system includes the option to make micropayments for individual searches or to gain full access to the archives as a bundle with a membership fee. The director of FT.com explained that the site "is all about giving the consumer a variety of options. We want to lock them into the subscription model as it is cash up front and easy to administrate, but the key is to offer them the choice." ³⁷

Choice and quality are key to micropayment’s success in the downloading of individual works. Differential viewing schemes offer great promise for micropayments if customers are guaranteed that they can rely on the quality and relevance of articles for which they are paying. With trusted systems in place one could envision a pay-per-print system, which eliminates the risk that makes consumers wary about paying for information the may not be able to use. Users could search to their heart’s content to find useful articles, and even read them online, but to print or save the file a viewer would have to make a 10 cent micropayment. Done in large volumes this could net significant returns for online publications. As more and more users turn to the Web as their primary source of news information one can expect a shift away from libraries and other traditional research venues to the convenience of a low cost alternative at home.

Of course anyone who has spent a minimum amount of time surfing the Web knows that advertising is the dominant revenue model for commercial sites. Nearly $1 billion will be spent on online ads in 1998, ranging from cars to cold medicine to porn. ³⁸ But that still does not mean that commercial sites are making money. According to Forrester the typical content-providing site will lose $2 million annually over next two years. ³⁹ Advertising will become more sophisticated as technology advances, with sites tailoring advertising to meet individual viewer demographics, but there is no evidence that better targeted ads will generate more revenue.

Forrester recommends that most content providers give up on user fees and concentrate on marketing demographic information instead. ⁴⁰ According to one report "The current off-line subscription model will be replaced by no-cost on-line subscriptions that require users to share personal data. Media companies will pay their bills by reselling this data." In other words the New York Times should not begin to charge for its Web version, but should sell profiles of its users to marketing firms and charge more to advertisers for being able to guarantee a certain audience type. ⁴¹ This approach could be enhanced by incorporating software into the site that reads a surfer’s demographic profile as she searches the Web and then directs the appropriate ad her way. ⁴²

Still, research indicates that branding is more effective through businesses’ own sites than advertising. ⁴³ Many merchandisers retain exclusive selling rights online, while most media providers will jealously guard their original content from other areas of the Web. Thus businesses should be willing to make every effort to attract Web surfers to their company sites.

One way to tie consumers in is to entice them with microcash credits for viewing a particular ad or site. Compaq is planning to use Digital’s Millicent micropayment technology to offer incentives for perusing its site’s ads. ⁴⁴ Browsers can receive small digital payments each time they visited a new advertiser’s site. These in turn can be used to pay for services on other partner sights. Alternatively, browsers could pay for the privilege of viewing ad-free sites by turning in some of their digital cash. As the new owner of Millicent, Compaq has an obvious incentive to employ the new system. But it could open a profitable channel of advertising revenue once electronic cash becomes more prevalent on the Web. ⁴⁵

The Compaq system presents both the best and worst case scenario’s for the future character of a Web dominated by micropayment technology. On the one hand browsers receive tangible rewards for their surfing in ways the regular economy can only simulate. On the other hand basic conveniences like ignoring ads or accessing interesting content are commoditized in ways consumers generally don’t contemplate. Although Cable and PBS each charge consumers for the privilege of ignoring ads it is done so on an indirect and monthly basis that consumers seem to find less offensive.

Most of micropayment’s potential lies years down the road, awaiting technical advances in encryption (enabling money to flow securely online), trusted systems, and bandwidth. As it now stands, almost all Web commerce is conducted with credit cards and are limited by the same minimums that apply to real world transactions. Micropayment pilot programs are hindered by cumbersome registration procedures and markets restricted to companies that have worked out prior contracts with microcash vendors. But technological fixes will undoubtedly emerge that allow vendors to protect their intellectual property from piracy and simple sharing. This will force consumers to contemplate what they find important enough to pay for and how they want to pay for it.

The examples illustrated here do offer some lessons for how and when micropayment technology could flourish. First, like most Web technology, micropayments rely heavily on network effects, and thus must wait until e-cash or some other form of digital money becomes widely used. In this sense Kelly and Goldhaber are right in that companies should concentrate on providing free services to grow networks. But their theories fail to see that once people begin to rely on networks different payment schemes can be imposed. Companies that adopt differential pricing schemes could choose to offer certain data to casual users for mere cents while charging businesses a much higher rate for the same information, as Shapiro and Varian predict.

Second, micropayments will not dominate the online commercial marketplace and will probably do little to stifle the free-content spirit of the web that is sustained by informational and personal sites. Rather, its greatest promise lies in conjunction with other payment plans like memberships or advertising revenue. By offering services on an individual basis, firms can create demand for bulk rates if customers’ only other option is to pay per view.

Finally, no micropayment system will succeed on a wide scale without trusted systems. Although advertising and merchandising applications like Compaq’s pilot could thrive on micropayment technology without copyright safeguards, most digital intellectual property will remain valueless until it is scarce. If trusted systems become prevalent however, a range of micropayment applications opens up. Vendors would no longer be limited to traditional product bundles. Individual publishers could reach vast audiences with cheap, original content that users could enjoy for fractional costs. And in the recording industry songs and videos could be offered individually for fractions of the current cost. Micropayments will play a key role in fueling this growth of cheaply available intellectual property.

Micropayment’s ultimate benefit will be better content for consumers. No matter how much free content businesses give away to attract attention, they will always have an incentive to hold something back for paying customers. But with Web volume and market size growing exponentially, companies will realize a strong incentive to lower prices, desegregate goods, and sell them to a massive microcash audience that they can reach almost for free.

FOOTNOTES

1. www.nytimes.com

2. www.weather.com

7. www.slate.com

9. Many sites combine these models. The Financial Times, for example, has just introduced a means of receiving content from its site via micropayments and subscription as well as placing advertising throughout the site. www.ft.com. For a good summary of how businesses realize value from their Web sites, see Chris Charron, Bill Bass, Cameron O’Connor, and Jill Aldort, Making Users Pay, The Forrester Report, 2:11, July 1998.

16. Free research is available at www.askmerrill .com.

24. David Weisman, Digital’s Micropayment Solution, Forrester Money and Technology Strategies, 2:15, March 17, 1997. Millicent’s Web site is www.millicent.com.

35. Nexis is the epitome of this model, charging exorbitant fees to businesses that can afford the service. Economist at www.economist.com and Wall Street Journal at www.wsj.com are some online examples.

36. Its Website is at www.ft.com.

44. Adam Creed, Compaq Resurrects MilliCent Online Payments Model, Newsbytes, November 12, 1998. The system will be tested with Compaq’s online journal, The Rapidly Changing Face of Computing, at www.millicent.digital.com/buy/rfoc.html.