The last time a Democrat controlled the White House, he defended globalization. He understood that trade creates losers, and he sought to assist them with progressive tax and spending policies. But today's leading Democrats are flirting with the opposite approach. They are forthright in their complaints about trade. But, with the notable exception of John Edwards, they are surprisingly reluctant to propose tax hikes.

You can see why Democratic sentiment has shifted. Globalization is no longer what it was when Bill Clinton shepherded the North American Free Trade Agreement and the Uruguay Round through Congress. China's economy has quadrupled in size; India has emerged as a competitor; the sense of insecurity has spread from a handful of vulnerable industries to much of the economy. And yet if they reverse the Clinton formula, the Democrats will be doubly wrong. They will be swallowing left-wing propaganda about trade and right-wing propaganda about tax cuts.

Paradoxically, the changes that have made globalization less popular have rendered resistance to it less fruitful. Back in the 1980s, trade put pressure on big, vertically integrated industries: cars, electronics. In the new world of outsourcing and global supply chains, vertically integrated enterprises have been sliced into discreet processes; trade now puts pressure on tasks rather than on industries. Back-office administration and phone-based customer support may shift to India, and this shift may affect industries from banking to medical services. The manufacturing and assembly of components may be outsourced to Mexico or Asia, and this change may affect everything from toys to telephones.

So trade now threatens workers in more industries. Even if it still causes less dislocation than technological change, we shouldn't be surprised that anti-globalization sentiment has sharpened. But the advent of competition in tasks also renders protectionist remedies less sensible than ever.

In the era of vertically integrated industries, there was a superficial plausibility in the economic-nationalist case for trade barriers. Big American corporations that did nearly all of their production at home were up against Japanese and German rivals that were equally rooted in their own countries: "If they win, we lose" was an understandable attitude. But now this picture has been scrambled. America Inc. had become Global Inc.: It's hard to erect trade barriers without disrupting corporate America's supply chains. The world's fastest-rising trader illustrates the point. More than half of Chinese exports are made by subsidiaries of foreign companies.

Advocates of trade barriers want to protect workers as well as corporations, but this rationale is also fraying. The old case for tariffs was based on the fact that trade-affected workers were often geographically clustered: If cheap Asian steel hammered a Midwestern steel town, protectionists argued that steelworkers could not be expected to adapt and find new jobs because steel was the sole prop of their communities. But the new competition in tasks rather than industries renders the tragedy of the company town increasingly rare. If a data-entry clerk loses her job in Omaha, she can probably find other work without having to move.

So trade protectionism, which always damaged economic growth, has now lost the few arguments that supported it. Globalization has caused understandable anxieties, but tariffs are a useless therapy. The alternative of helping globalization's losers with tax-financed spending programs is more promising by far. For, contrary to what conservatives routinely assert, tax-and-spend programs need not be bad for the economy.

Consider the work of Peter Lindert of the University of California at Davis. In a magisterial work published three years ago, Lindert analyzed tax-financed transfers across rich economies and found no correlation with the rate of growth or with gross domestic product per person. This, Lindert continued, should not be surprising. What matters for growth is less the quantity of tax and spending programs than their quality.

Lindert is no party-line liberal. He argues that high taxes in Europe don't damage growth because they hit consumption and labor rather than savings and capital: This is an uncomfortable point for those who want the tax system to be progressive. But Lindert also argues that high taxes are compatible with growth if the revenue is spent well. Investments in education and public infrastructure boost a country's growth rate. Programs that break the link between employment and health insurance enhance the flexibility of workers. Subsidized child care keeps women in the work force.

None of this suggests that high taxes are risk-free, especially if they finance programs with no growth payoff — such as entitlements for the elderly. But knee-jerk opposition to raising taxes is misguided.

The government could and should spend more on making healthcare portable and retraining displaced workers. That is a much better bet than protectionism.

Sebastian Mallaby is a member of The Washington Post's editorial page staff.