Imagine walking to the edge of a cliff, looking down, saying to yourself: "Boy, that's a steep drop. It'll sure hurt if I jump," and then jumping anyway. Such seems to be the curious strategy of the world auto industry in the late 1990s, as it grapples with prices that are actually declining, demand that's stagnant, and production capacity that, against all good sense, just keeps getting bigger.

Now, in one respect, this anxiety about the auto industry simply reflects the fact that it has become the poster child for a global economy in which deflation is suddenly everyone's favorite thing to worry about. Even in today's supposedly high-tech world, General Motors and Ford remain bellwethers for the way people feel about the economy as a whole. In the mid-1970s, Detroit's gas guzzlers were taken as emblems of the United States' inability to adjust to the end of the postwar boom. At the end of that decade, nothing made inflation more real than the phrase "sticker shock." So today the image of assembly lines churning out hundreds of thousands of unneeded and unwanted vehicles is an excellent representation of the specter of global recession. This is a specter that seems to loom larger, of course, after the series of currency crises that have beset Asia.

In another respect, though, concerns about global overcapacity in auto production are industry-specific and, by most accounts, well justified. It isn't just the William Greiders of the world who think that, in the near future, there are going to be too many cars and not enough drivers. Instead, just about everyone thinks this. What's curious is that no one is really doing anything about it.

It's true that overcapacity concerns are at least a decade old (click for a quick review) and that the predicted crisis of demand has not yet materialized, though perhaps it has, and I've just seen it as a pleasant period when car prices have stayed the same instead of rising. That's because the U.S. economy has continued to expand so briskly and because the Big Three automakers have restructured their operations to do a better job of matching supply with demand. (In other words, they downsized.) Overcapacity is, in any case, difficult to measure, since it's more important for car companies to be able to meet increases in demand than it is for them to make only as many cars as customers want. You can shut down a line for a couple of weeks without any problems if demand falls short. You can't build that line from scratch if demand rises unexpectedly. By some estimates, in fact, overcapacity of a million cars is reasonable for the Big Three.

Y ou wouldn't be wrong, then, to see something a tad Chicken Littleish in all this overcapacity talk, especially when it's being proffered by the very people building the cars for which there is supposedly no demand. And there is clearly an element of saber rattling on the part of U.S. car companies, which are anxious about the continued impact on their earnings of the strong dollar and who want South Korea, in particular, to open its markets to U.S. cars. Last year, the Big Three sold 3,900 vehicles in South Korea, out of a total market of 1.6 million. A Pontiac Bonneville is as rare a sight in Seoul as an Aston-Martin is here. Emphasizing the global realities of overexpansion thus becomes a way for the Big Three to serve their own interests without seeming parochial.

And yet there are those realities, and South Korea is the most remarkable of them. At a time when the South Korean economy as a whole seems to be in need of dramatic restructuring precisely around the issue of excessively free credit for capital expenditures, South Korean car companies are forging ahead with investment plans that dwarf anything they've previously done. As a whole, South Korean automakers plan to put $3.72 billion into expanding facilities. Hyundai--formerly the Yugo's poor cousin--is increasing capital spending by 700 percent. Daewoo is doubling its investment. And, most incredibly, Kia Motors, which needed a government bailout to save it from its creditors, will be putting $500 million into plant expansion in 1998, a threefold increase. Oh, and then there's Samsung, maker of cool stereos and microwaves, which decided last year that it wouldn't be satisfied until it had become a world-class automaker. What about hand-tooled leather bags? They're nice. Couldn't Samsung have decided to become a world-class maker of hand-tooled leather bags instead?

Similarly, Mexico's output--two-thirds of which will be exported--is expected to rise 10 percent a year for the next three years, while even the Japanese, whose car companies are among the only bright spots in a forlorn economy, are adding to capacity. And, of course, U.S. automakers, even as they worry about their inability to raise prices, are breaking ground on new plants and pushing their workers to take more and more overtime.

In that sense, South Korea is just a striking example of what's happening all over. And it is hard, when reading yet another account of an auto company CEO's concerns about falling prices, not to want to shake him vigorously and say, "Just say no, damn it!" If there are going to be too many cars, then it seems sensible to stop making cars. Except, of course, that if you do you lose all hope of capturing what market does exist, and except, of course, that when you stop making cars, you have to close factories and throw people out of work.

It's important to see that the problem here is not, no matter what General Motors says, that U.S. workers are overpaid. Nor is it really that individual factories are overstaffed. After all, the Big Three are enjoying hefty profits and spending billions on share-buyback plans. All three companies require between 3.27 and 3.56 workers to build a vehicle--significantly inferior to Japanese standards but significantly better than it ever was, and significantly more efficient than most people thought possible for a U.S. company. (And no, I don't know what a 0.27 worker looks like.) The problems that U.S. automakers endured during the 1970s and 1980s were, for the most part, self-inflicted, as the superior performance of Japanese automakers over that period suggests. But the problem that U.S. automakers face now is of a macroeconomic, not microeconomic, nature, which is what must make it so difficult for them to accept. It's about the fact that within a given industry, supply does not ensure demand, and that what is rational for the individual firm--which needs to sell as much of its product as possible and so has to make sure that its product is available--may be collectively irrational.

One interesting question raised by all this is whether cars are commodities. In other words, does the fact that there are 2 million more Kias on the road really make someone less likely to shell out $22,000 for a Honda Accord? Auto design and advertising are all about differentiating one product from another. But all this talk about global overcapacity suggests that automakers see themselves as all making the very same thing. I wonder if that'll make me feel better the next time I see someone blow by me in a new Porsche Boxster.