Thirty-five years ago, Barbra Streisand made her Broadway debut in a musical called "I Can Get it for You Wholesale," which wasn't about electronic commerce but perhaps should have been. The title came from the ability that some people had to buy things for a much lower price than you or I could get. The only way to get such a low price normally was to buy whatever was for sale — mink coats, cotton balls, coal — in large amounts, thereby getting a volume discount. Such wholesalers then theoretically resold the same goods at a higher retail price. But some lucky people were able to buy small volumes at the large volume price, to get the stuff at wholesale prices. You had to know somebody to buy at wholesale. That was the trick.

There is a certain logic to this wholesale/retail system. The idea is that there are fixed costs associated with making a transaction and it costs just as much to keep a small buyer on the books as a big buyer. Dividing these fixed costs per widget sold, it costs a lot less per widget to sell a million of the things than to sell a thousand or ten. So the volume buyer got some of that cost differential back in the form of a lower purchase price. That's the way it is supposed to work, but the essence of the musical was that there were people who mysteriously and magically knew how to get the wholesale price without buying large quantities. In the musical it usually had to do with being someone's brother-in-law, which is the way it still works in some places.

The sad truth is that some people just know how to get the same thing for less money. I have never been one of those people, though.

The rules have changed a bit since 1963. Discount retailers like Sam Walton figured out that by offering wholesale prices to retail customers they could destroy smaller competitors and gain market share. Get enough market share and you can pretty much set your own price with suppliers, which is what Wal-Mart does, creating one of the world's great fortunes in the process. Wal-Mart flipped the pricing algorithm primarily by using computer technology to drive the fixed cost of serving an individual customer so low that it no longer even mattered.

On Wall Street, Charles Schwab invented the discount brokerage business because computers allowed him to service an individual brokerage account for almost nothing. Like Walton, Schwab went for volume.

In both of these examples, information technology made possible the change in consumer pricing, but the actual commercial venue didn't change. Wal-Mart still built stores and Schwab brokers still answered their telephones. But now we have the Internet, and for the first time the technology that runs the back-office also runs the storefront. In electronic commerce, the entire transaction takes place inside the network. This is good, right?

Yes and no. Some very smart people think eCommerce, for all its obvious advantages, might actually destroy commerce rather than helping it.

Before we get into an attempt to understand the psychology of electronic commerce, we need to define two terms: "commodity" and "brand." A commodity is a product that is so standardized that it doesn't matter where you buy it. Number two heating oil, if it meets the standard that defines number two heating oil, is exactly the same no matter where you buy it. Only the price, transportation distance, and available volumes vary. So too with winter wheat and extra fancy cranberries and a thousand other products you can buy by the ton. A brand, on the other hand, is supposed to be the functional opposite of a commodity. A Louis Vuitton handbag isn't just a handbag, it's a Louis Vuitton. Consumers will generally pay more for a name brand because it signifies some extra quality. There is also the idea that some name brands are available in only small quantities, which is why Rolls-Royce cars are supposed to cost so much.

In the 1990s, however, we've managed through our ability to manufacture the heck out of stuff to effectively turn some name brands into commodities. Sony makes great TV sets, for example, but do we imagine for a moment that they won't make enough of some model to meet demand? Of course not. Sony will make as many TVs as they can sell and every TV will be just as good as the others. Sony TVs have become a commodity, which means the only real difference between buying the same TV at one store or another is price.

Now we finally get to the Internet, which is turning into the greatest comparison shopping engine in the history of commerce. This is because the Internet eliminates from shopping the effects of time, distance, and effort. In the past, maybe you could get a better deal on that Sony TV from a dealer across town or across the state, but you first had to find that dealer. As the Internet grows as a marketplace in its own right it becomes effortless to find the best deal. This applies to any mass-produced brand-name item. Oracle CEO Larry Ellison, for example, buys his underwear over the Internet because he knows what brand and size to buy — Munsingwear briefs size 34 with a kangaroo pouch — so why pay more?

Why pay more, indeed. But where does that leave the retailer? In trouble, that's where, even when that retailer is on the Internet. Especially when that retailer is on the Internet.

Let's do a simulation. We have a manufacturer, Sony, that will make as many truly great color TVs as the world requires. Then we have two Sony dealers operating on the Internet. Let's pretend these two dealers are the only Sony TV dealers in the world and that they sell everything over the 'Net. What happens to TV prices? Well it's possible, I suppose, that there could develop a stable market with each dealer taking half the business or maybe one taking slightly more or slightly less. But for that to happen there would really need to be an agreement between the two dealers. In America we have anti-trust laws precisely to avoid such agreements and thereby protect consumers. Of course, as Microsoft wants us to remember, there was no Internet when those anti-trust laws were written.

Absent an illegal pricing agreement, then, what happens to our two Sony dealers? According to a mess of academic studies, the dealers fall into a price war that ends only with the death of the weaker dealer. Rerun the simulation with many dealers and it ends the same way with bodies all over the place. Make the Internet a bigger part of the total market and the bloodbath just happens faster.

One thing that's especially odd about this behavior is that nobody seems to be freaked by it. Papers about just this eventuality have been published by researchers at IBM and several universities, yet nobody is talking about the perils of highly-efficient markets. Maybe it's nothing to worry about. Or maybe it's just too early. Everybody yawned when I wrote about the Y2K bug back in 1991, but they aren't yawning now.

This is a very real problem, but a simulation doesn't completely define reality. The whole eCommerce experience is likely to be much richer than just dueling Sony dealers. And I suspect there may be certain attributes of this emerging market that will take us all in directions we can't even imagine right now. But that's for next week's column.