The argument that we are living, or at least entering, a new economy rests on two legs: the spread of computers and the ongoing systematic organization of the world economy. While the first point gets most of the press, the latter may be almost as important. Small, locally defined markets sacrifice value by limiting both economies of scale and the growth of specialization. As markets grow globally, the resulting increases in efficiency should release a continuous stream of value into the world economy.

It seems that the hottest piece of the IT market would be where these "legs" join, with the use of computers to manage cross-border transactions. An article we ran in 1993 ("Made for Export," Dec. 15), though narrowly focused on the migration of a single major corporation’s export management system from paper to software, seemed to catch this spirit. According to our piece, this transition had no downside to speak of: You just wrote the software, booted her up and counted the dollars.

Given this, it seems paradoxical to learn from those involved with cross-border trade that the field was one of the last corporate functions to be digitized. "You could see typewriters in offices into the late ’90s," says Cris Arens, president of Fountainhead International, a logistics software house in Bensenville, Ill. "And they were being used too."

Arens and other sources offer many reasons why cross-border logistics were so slow to adopt IT. Over the decades, international trade had built up a huge paper infrastructure, involving countless forms issued by public and private bureaucracies around the world (packers, inland carriers, international carriers, banks, insurers, freight forwarders, warehousers, customs and more), each with its own data and format requirements. Digitizing enough of these forms to make the transition useful would have been a huge job, and some data had to be on paper anyway, as a matter of law. For most companies, cross-border trade was essentially a revenue afterthought, which meant the resources to pay for the shift were hard to find.

Finally, there was no overwhelming reason to change. The high fixed costs of each transaction and relatively slow business cycles encouraged trade to move in small numbers of large shipments?one 10,000-unit shipment might arrive in February and then another in the fall?and typewriters could keep pace with that flow.

Through the ’90s, however, governments around the world launched a number of initiatives to lower trade barriers. It would seem common sense that freer trade would translate into a simpler process, but that is not how trade adjustments work. Trade is a highly charged process politically; changes always involve extended haggling over the finest possible details. The end result is usually a complicated stew of solicitation and qualification instruments, and refund claims tied to detailed lists of specific products and country-of-origin content documents, all linked to phase-in and phase-out schedules that themselves are constantly changing.