If the wishes of federal and state economic development officials all came true, U.S. industry would be a high-tech, white-collar paradise. Billions of dollars in subsidies, municipal land-use policies strikingly skewed toward IT business, and the unprecedented state sales-tax moratorium Congress recently bestowed on the Internet show how profoundly our government has bet on a dot-com future. But that narrowly defined future can -- and likely will -- contract.

For their part, new-economy entrepreneurs, on whom regulatory affection is being lavished, aren't complaining. "New York," gushed a high-tech lobbyist this spring, when the city expanded its Internet subsidy program, "is standing out on the information highway and showing a little leg."

It's easy to see why. Since 1993, new-economy enterprises have allowed the richest 1.5 percent of U.S. citizens to pocket a whopping $600 billion of the nation's $1.2 trillion taxable income growth -- one of the most impressive high-end wealth increases in history. The subsequent flood of capital gains and income tax revenues rescued virtually every public budget in the country, and government fell hard for anything Internet.

But, even now, with evidence of the possible pitfalls of IT emerging, obsessed public policy is apparent. The economies of regions that foster primarily white-collar service and technology jobs grow far more slowly than those that support a full complement of industries. Moreover, when elite IT employment crowds out everything else, overall educational performance and income distribution suffer.

Why, for example, should Latino K-12 students in Texas dramatically outperform their California counterparts on standardized tests? There are many cultural and political explanations for this, but it's quite possible that the most important factor is that Texas generates the nation's widest range of working-, middle- and upper-class employment opportunities. California's relatively tepid growth, in contrast, is heavily dominated by high-end service sectors. Low-income Latinos living in cities such as Houston and Dallas have far more opportunities to support their children's education than those in high tech-infatuated San Jose, Calif., or Los Angeles.

The Internet and high-tech growth seem to generate elitist politics. Today's new rich are quick to adopt the conceits typical of such privileged groups as Hollywood actors. When high tech turns entrepreneurs into new millionaires, they quickly become avid advocates of quality-of-life politics, espousing slow growth and other exclusionist policies.

These sentiments predominate in such new wealth centers as San Francisco; Seattle; Austin, Texas; Raleigh-Durham, N.C.; and even Manhattan. And broad-spectrum economic development is almost always the primary casualty. With constraints on the supply and development of commercial, industrial and residential land, rents and business expenses skyrocket. Unsubsidized, "old-economy" sectors -- especially blue-collar manufacturing -- get chased into more hospitable, less costly regions, if not overseas.

Mesmerized by Silicon Valley's success, few economic planners care that the price of high-tech development will likely be a narrow industrial base, massive wealth differentials and substantially slower growth. But they should. To thrive in the long term, IT, like all other industries, needs to interact and grow with the rest of the economy.

Consider, for example, the recent productivity boom. Since 1995, U.S. output per labor hour jumped suddenly after more than two decades of stagnation. Faster productivity growth is, of course, a very good thing. It stimulates rapid expansion, coupled with low inflation and unemployment.

Most economists and financial advisers attribute that improvement in performance to the diffusion of IT throughout the economy. But according to Northwestern University economist Robert J. Gordon, there's scant evidence that technology is leading the way. All the productivity gains, in fact, have occurred in durable-goods manufacturing, which comprises only 12 percent of the nation's economy; computer production accounts for the lion's share of that growth. At the same time, productivity actually has declined in the nearly 90 percent of our economy that is not making durable goods.

The data suggest that during the 1990s, America's focus on a narrow range of policy-preferred sectors artificially segregated IT businesses from the rest of the economy. Trouble is, the overwhelming majority of companies and industries that have not yet joined in the nation's productivity boom comprise the natural market for IT. It is shortsighted indeed to foster an economy that builds up novel technical capacities and solutions but simultaneously reduces the customer base for such products.

All too soon, our giddy public officials may taste the downside of the unbalanced economy they've helped build. If stock markets don't sharply recover, the capital-gains windfalls now feeding public budgets will evaporate. The lesson should be obvious. Government and IT sectors alike should expand our industrial options and potential markets for the new economy. Now, while we can afford to do so.