In one of the more startling public statements by a banker in 1996, Citicorp chairman John Reed said it would take two generations - 50 to 70 years - for on-line electronic banking to gain full public acceptance.

Taken out of context, his remarks to a Treasury Department conference on electronic money sounded like an invitation to complacency, or a dose of disinformation from one of the world's more aggressive purveyors of electronic financial services. But Mr. Reed chose his words carefully, citing a lesson learned from his 30-plus years at Citicorp: Banking markets, and society generally, take time to change.

He seemed to suggest that high-technology advocates can become so enthralled with the elegance of their systems and convinced of their viability that they overlook the most common of all constraints: consumer behavior.

"Privacy and security are at the top of the list" of consumers' concerns, the Citicorp chief executive said. "They won't deal with anyone who doesn't give them assurance."

While "some early innovators will be your electronic banking customers," he said, "the average consumer is not there yet and isn't going to be there" for some time.

"This is not a question of economics or efficiency. It is a question of trust. The consumer will have to trust you. The Internet is fundamentally flawed in that regard."

Essentially alone among the major U.S. banking organizations, Citicorp has been openly wary of Internet security and refrained from joining the rush to interactive banking and monetary transactions via the World Wide Web.

Mr. Reed and his senior technology officer, Colin Crook, have publicly expressed interest in and enthusiasm for the Web but not yet for transactional purposes. When Mr. Reed was asked during the Treasury conference in September when Citi would offer Internet banking, he replied, "Not until it's secure."

"There is no absolute security," said Mr. Crook, perhaps the only banker raising concerns about an "information warfare" attack on the banking system. "It is a risk management issue."

The Citibankers contend the risks of cyberspace are fundamentally different from those in other payment systems, and have yet to be addressed. "Security will be more demanding than even the government itself is used to," Mr. Crook said at the Treasury meeting.

Cynics are quick to note that Citicorp and Citibank are doing just fine off the Web. The bank serves more customers via personal computer than any other, through conventional dial-up connections and with software it developed more than a decade ago.

Citibank also has placed a bet on a digital currency for on-line transactions, the invention of one of its own vice presidents, Sholom Rosen. The bank claims it will be more secure than competing alternatives like Cybercash Inc.'s Cybercoins, Digicash Inc.'s Ecash, and the Mondex smart-card-based system.

Putting considerable prestige and intellectual firepower behind its cautionary principles, and behind the notion that the issuing of electronic currency should be reserved for regulated financial institutions, Citicorp has kept alive a debate that is likely to resound for months if not years in public policy circles, with effects not just on the battle for technical and competitive superiority but on the very consumer behavior Mr. Reed is trying to gauge.

Consider some recent twists and turns:

The U.S. government continues to struggle toward a policy on data encryption, the technology crucial to on-line transaction security, that would be agreeable to the high-tech community while addressing national security and law enforcement concerns.

A May 1996 report by the National Research Council of the National Academy of Sciences - Citicorp participated in and vocally endorsed the study - criticized the government for being backward with its restrictions on encryption, particularly regarding its export. (See related article on page 14A.)

Hewlett-Packard Co. in November announced its International Cryptography Framework, the first "strong encryption" method to get U.S. export clearance. While the framework is adaptable to various and changing government policies, it did not fully resolve the controversial issue of access to encryption keys.

An information security team at the National Security Agency produced a monograph (excerpted at left) critical of the degree of anonymity built into Digicash's Ecash. The NSA, of course, is part of the establishment attacked in the National Research Council report.

Digicash and Mondex, which is being taken over by MasterCard International, continually trade charges about their degrees of anonymity and security. Both sell anonymity of payments as a necessary analogue to cash. In that Digicash's anonymity appears more absolute, it may raise more governmental concerns. But Digicash, the brainchild of the renowned cryptologist David Chaum, accuses Mondex of not being "true electronic cash."

First Virtual Holdings Inc., an Internet payment pioneer, does not trust Web security; its transaction data flow instead over private E-mail. By contrast, Cybercash Inc. chairman William Melton is so confident of the available technology that he tells bankers: "Security is essentially done. Just tell your customers, 'Don't worry, we'll take care of it.' " (He is more worried about privacy as a political flashpoint.)

Enter the central banks of the Group of 10 industrialized countries, the constituents of the Bank for International Settlements in Basel, Switzerland. This august global regulatory body has signed off on a moderate, largely laissez-faire approach to the electronic evolution of money.

A task force empaneled by the G-10's payment and settlement systems committee, which is headed by Federal Reserve Bank of New York president William McDonough, spelled out its conclusions in a 64-page booklet, "Security of Electronic Money," dated August 1996. (See excerpts beginning on page 7A.) The task force was generally impressed by existing security capabilities, particularly those incorporating hardware components like smart cards.

The report took the eight-member task force less than a year to complete. Chairman Israel Sendrovic, the New York Fed's executive vice president of automation and systems services, asserts that this was no rush to judgment. He personally did due diligence on all of what he calls "the usual suspects" - the electronic money schemes not mentioned by name in his report (but presumably in this article).

In a recent interview, Mr. Sendrovic stressed that there are no absolutes. "There is no such thing as one secure measure," he said. "It's a combination of measures, and the combination of measures changes the risk management of an attack."

His measured response to a lot of questions - pertaining to money laundering or the market potential of electronic currency and how it is to be regulated - was, "It depends." He did say, in response to the recent flurry of questions about smart card security emanating from Bellcore and other research laboratories, that the cards were advertised as "tamper- resistant, not tamper-proof."

Mr. Sendrovic said his panel has disbanded, satisfied with its work and having gotten positive feedback. "Then again, it didn't break new ground," he said. "Remember, it was designed not for the cognoscenti but for the Group of 10 governors.

"We stay in close touch and follow these things," he said of the task force, adding that it may have cause to renew its inquiry in a year or two.

Though the task force acknowledged "comprehensive security risk assessments of the entire system" are still lacking, it said they are within reach. And its words lacked the alarm or urgency of, say, the Citicorp contingent.

Sholom Rosen, inventor of Citibank's Electronic Monetary System, characterized the risks as "very high" and not yet fully analyzed. Digital cash gains legitimacy when it is interchangeable with other forms of money, he said, but its interactions with those systems - how an attack on one mechanism would affect others - must be studied. And he said he believes the answers do not lie in technology alone but in the fundamentals of the "three pillars of security" - prevention, detection, and containment.

Where Mr. Rosen sees enormous hazard, Mr. Sendrovic retains faith in barriers to entry, as might be expected of someone who has worked with the dependable Fed Wire for many years. To be legitimate, electronic money "has to be cleared," he said. "At some point it has to get into the payment system."

Is "the payment system" at risk of infection from the new forms of money? Based on what we know so far, it depends.

Subscribe Now

Access to authoritative analysis and perspective and our data-driven report series.

14-Day Free Trial

No credit card required. Complete access to articles, breaking news and industry data.