"Cash Poor": E-Cash 정착이 어려운 이유

Cash poor
The Economist
London
May 10, 1997

Just as computers need to convert the world's complexity into digital bits before they can make sense of it, futuristic digital markets need a correspondingly electronic currency - something as liquid and simple as cash, yet as international and transportable as a credit-card number. The solution is digital money - electronic tokens so protected by encryption and other safeguards that they can be accepted online with all the confidence of real cash. By 1995, there were at least 7 competing payment schemes, from firms such as CyberCash and First Virtual. However, digital cash has few takers as yet. The trouble is rooted in a fundamental error. The digital-money business is built on the premise that electronic commerce will bring a boom in microtransactions for information goods or services. The psychology of it is all wrong.

Individual consumers dislike paying for information and positively hate
meters. Ultimately, the biggest problem for digital cash is that there
is nothing very wrong with the traditional ways of paying for things.

Copyright Economist Newspaper Group, Incorporated May 10, 1997

Full Text:

THE geeks love electronic commerce beyond all reason. It is not that they
are more moneyminded than others, or more interested in shop ping: it is
simply the technology. Online markets are a chance to wheel out every cool shopping simulation ever invented in a computer science laboratory. Imagine recreating an entire payment system in software. Add "autonomous intelligent agents" to cruise the Internet, shopping for you, even waving your bid in "virtual auctions". Then assume that the goods bought and sold are information in all its varieties, shipped digitally down the same networks. It is a programmer's dream: computers running the world's online marketplaces on behalf of their human masters.

Just as computers need to convert the world's complexity into digital bits before they can make sense of it, these futuristic digital markets need a correspondingly electronic currency: something as liquid and simple as cash, yet as international and transportable as a credit-card number. The solution: digital money-electronic tokens so protected by encryption and other safeguards that they can be accepted online with all the confidence of real cash.

The boffins jumped at it. Even before the first Web shop had made its first sale, new e-money companies were devising suitably wired ways for consumers to pay. By 1995 there were at least seven competing payment schemes, from firms such as CyberCash, Wave Systems, First Virtual, DigiCash and Digital Equipment, along with several universities. By 1996, the first three had gone public at a combined valuation of $1.4 billion. Each firm's technology was mathematically brilliant; most were built on encryption so elegant, so fascinating, that no one bothered to ask: who needs it?

Answer: not many, at least not yet. For all the beauty of digital cash,
it has won precious few takers. A small number of shops around the world
will accept some form of digital cash as an experiment, but hardly anyone
holds any. The three listed companies combined have done less than $3m-worth of business in their entire history; not coincidentally, their total valuation has fallen by over $i.2 billion since going public.

Their troubles are rooted in a fundamental error. The digital-money business is built on the premise that electronic commerce will bring a boom in "microtransactions" for information goods or services: buying a newspaper article for a quarter, say, paying a few cents a day to see a favourite comic strip, or renting software by the hour. Consumers will love it, the boffins insist. They can pick and choose what they want from the global marketplace of ideas, without committing themselves to expensive subscriptions. If you assume such a market in tiny one-off transactions, then digital cash makes a lot of sense. Small credit-card purchases are uneconomic to process, and a quarter is hard to shove down a modem.

Pennies for your thoughts

Such digital micropayments promised to democratise commerce: anybody with
an idea, a piece of news, anything at all that someone, somewhere, might
want to see, could become a publisher online. No need to assemble a constant stream of content to justify a subscription or attract advertisers: just put it on the Web and take digital pennies for your thoughts. A popular journalist might make a fortune by selling stories direct to readers. The journalists, naturally, did the calculation: 250,000 readers at, say, 25 cents equals $62,500 a day, enough to test the loyalty of any reporter. No wonder the press went wild.

But the hacks may need to stick with their day jobs after all. Unfortunately for both such would-be infopreneurs and for the digitalcash companies, the psychology of microtransactions is all wrong. Individual consumers, the most likely takers for such a la carte fare, dislike paying for information, and positively hate meters. Witness the poor reception of pay-per-view television and the enthusiasm for flatfee Internet access. Magazines and newspapers sold on the newsstands are subsidised by advertising to bring down their cost; broadcast television in most countries is entirely subsidised.

Consumers want to avoid wasting their money, but more importantly, they
want to minimise their risk. They choose well-known brands to avoid disappointment, and are attracted by package deals that promise no hidden costs. Yet few things are as risky as, say, a teenager (the most technologically savvy yet financially irresponsible member of the household) let loose on a service that charges by the minute. The trouble with metered goods is the infinite downside, the vast bill at the end of the month or the unexpectedly empty account.

Sometimes this leads to counter-intuitive behaviour, such as consumers
preferring a flat all-youcan-eat fee when pay-as-you-go would be cheaper.
In an unpublished paper debunking many of the economic assumptions of electronic commerce, Andrew Odlyzko, an AT&T' researcher, gives the example of a 1970s experiment by the then Bell System to offer metered local telephone calling in addition to America's usual flat rates. Metered service was $5 a month for so free calls, after which calls were 5 cents each. Flat-rate was $7.50 for unlimited calls. Studies found that 50% of the customers who were making hardly any calls at all, and thus would have saved money by switching to metered service, nevertheless stayed with flat rates.

Eric Hughes, a cryptology expert with Simple Access, an electronic
commerce consulting firm, conducted his own experiment: at a technical conference, he issued a roll of pennies to every participant and explained that there was a one-cent charge, payable into a box beside the door, each time anyone entered the main room. Some of the participants dumped their entire roll in the box at the first opportunity and declared themselves issued with a free entry pass. It was just easier that way: nobody likes to be reminded of the cost of each transaction.

Content providers do not like payments for microtransactions either, because they punish avid users, nickel-and-diming them to death, while rewarding dabblers by giving them what they want at low cost. As Nathan Myhrvold, Microsoft's chief technology officer, puts it, "This is called `screwing your best customer.' "

Writing in Slate, Microsoft's online magazine, Mr Myhrvold explains that
the main problem with usage-based charging for electronic information is
that it is an artificial distortion of the underlying economics. Once an
article is on a Web site, it costs the publisher nothing to have it read.
So why charge as if it did? One solution might be to offer a variety of
payment plans, to maximise the number of readers: eager ones might go for
a subscription, while browsers might want to pay just for the occasional
article. But there is a better way to maximise the number of readers, Mr
Myhrvold argues: drop the price to zero and make your money from advertising. And that, indeed, is what most online publications have done, to the dismay of the digital-cash companies. "The problem is that the new payment mechanisms matured before the market did," says Mr Hughes. "The real competition in informational goods isn't other payment schemes, but advertising."

But some would-be publishers will never have audiences big enough to attract advertising, or loyal enough to buy subscriptions. What should they do? The sex industry, ever innovative, has proved a trailblazer in the economics of new technologies. There are thousands of individual porn sites on the Internet, many of them offering access to their wares for around $l0 a month. Punters are rightly dubious about the quality (and reliability) of such ventures. So a firm called AgeCheck, under the guise of certifying that the consumer is an adult, sells a subscription that covers dozens of them. The consumer's risk is reduced because it is distributed over many sites, and the sites get a proportional cut of the subscriptions. A bit too racy a model? Think of a day pass at Disneyland instead. Small publishers can band together, perhaps by theme, and sell one subscription for all.

Even a much-touted advantage of digital money-that it is anonymous, just
like real cash-is leaving the consumers cold. The flip side of anonymity
is risk, because the currencies that claim this feature-mainly DigiCash
and Mondex-do not confirm each transaction with a central bank. If a hacker were able to counterfeit the script, the entire currency might become worthless. Moreover, not many consumers are bothered about anonymity: people tend not to value their privacy until it has been invaded, and the Internet is too new to have thrown up many such abuses yet. Studies in realworld commerce suggest that consumers are willing to pay a 3-t0% premium for anonymity in transactions, but no more, says Mr Hughes. For highvalue products it may be less: as DMG'S Mr Gurley points out, customers like to know that the manufacturer or retailer has proof of the purchase, if only to make after-sales service easier.

Tested, not tired

Ultimately, the biggest problem for digital cash is that there is nothing
very wrong with the traditional ways of paying for things. For hard goods, credit cards are fine. Thanks to some basic encryption technology (which will be much improved with the forthcoming SET standard-see box on previous page), they are relatively safe. And the card firms are eager to increase their reach to make them the de facto coinage of cyberspace. "What used to be a $20 minimum purchase appears to be falling lower and lower every day," writes Mr Gurley. "Currently most credit-card companies will support transactions as low as $6, and we suspect that that number will be dramatically lower in 612 months. Micro-payments may be fighting an uphill battle against some extremely frightened, yet powerful competitors."

Likewise, most of the physical world's economics of selling information-advertising, subscriptions, and bundling-have moved online with surprising grace. None of them faces the difficult problem of persuading both consumers and merchants to adopt a radically new financial system (credit cards and cash machines took more than a decade to gain wide acceptance).

But none of this means that digital cash will never take off. Old habits
always bridge new technologies: Alexander Graham Bell thought the telephone would be a good way to broadcast concerts. Yet in these early days, the chicken-and-egg problem looms large. The irresistible businesses that can work only with digital cash will not arrive unless enough people have digital cash to spend, but few people are willing to lock up good money in electronic tokens with nowhere to spend them.

To succeed, digital cash needs a bridge between the old world and the new. The closest to that is First Virtual's "virtual PIN", which is a way of aggregating small transactions into a monthly credit-card charge, using nothing more high-tech than an email confirmation with the buyer to avoid fraud. But after three years of limited success, the company is still trying to fine-tune its model: shortening the time it takes for the merchant to be paid, and persuading its credit-card company backers to seed the market by pitching the service to their millions of card-holders. Even then, the First Virtual system does not offer all the advantages of real digital cash: its fees are too high for truly tiny sales (29 cents for each transaction, plus 2% of its total value)unless the merchant himself goes to the trouble of combining many little purchases. Moreover, its encryptionfree process is too vulnerable to fraud to be suitable for hard goods that impose a real cost on the merchant, such as books.

To take off, digital cash will probably have to work hand in hand with
similar electronic money tokens in the physical world. Companies such as
MasterCard's Mondex subsidiary and Visa International are promoting "smart cards" with stored value, similar to European and Japanese telephone cards but for use anywhere. Credit cards started as loyalty cards for individual chains of retailers; the same may happen with digital cash cards. Subway cards could be used to pay for coffee, and telephone cards might also feed parking meters. Thousands of merchants around the world are participating in trials of generic cash cards of various sorts. Moving all these cards online will be a natural evolution. The winning standards will be those that fare as well in the physical world as the virtual.