The Bush administration’s decision last March to impose tariffs of 8-30 percent on steel has been called everything from hypocrisy and stupidity to Machiavellian political brilliance. The reaction has been a remarkable demonstration of the strength of free-trade opinion in the United States -- but it has also been a bit puzzling.

The steel tariffs, even if one believes they are bad policy, are just temporary aberrations from the norm; they will be lifted in a couple of years. But for dozens of other products -- sneakers, spoons, bicycles, underwear, suitcases, drinking glasses, T-shirts, plates, and more -- tariffs of 8-30 percent are neither aberrant nor temporary. In fact, they are normal and permanent parts of U.S. trade policy. Barring a deliberate change in policy, they will never be lifted -- and no one seems to care.

The reason is not simply that people care more about steel than about underwear. Rather, it is that the tariff system has become an obscure, little-studied topic. Those who debate trade and globalization view tariff policy as boring and out of date. Career trade negotiators who set tariff rates have little contact with the customs officers who collect the money. Journalists cover political debate and international disputes rather than the functioning of permanent policy. And government officials, aware that tariffs are generally low and raise little money compared to domestic taxes, rarely think about the system as a whole.

But if these groups were to look more closely, they would find a remarkable situation. Tariff policy, without any deliberate intent, has evolved into something astonishingly tough on the poor. Young single mothers buying cheap clothes and shoes now pay tariff rates five to ten times higher than middle-class or rich families pay in elite stores. Very poor countries such as Cambodia or Bangladesh face tariffs 15 times those applied to wealthy nations and oil exporters. Despite this dismal situation, however, fixing the system would be easier than many imagine.


The problem arises more from neglect than from malice. No U.S. administration since the 1970s, or perhaps even since John Kennedy’s, has had a specific vision for tariff policy. Regardless of party, administrations have instead seen tariffs as a series of discrete issues that are useful in building domestic support for, and coopting potential opposition to, larger trade agreements.

In past trade negotiations, some domestic interests -- manufacturers of semiconductors, chemicals, capital goods, and so on -- sought export opportunities by advocating the elimination of overseas trade barriers and were willing to give up tariffs at home in exchange. Other industries -- shoes, textiles, cutlery, glassware -- feared foreign competition and fought to keep tariffs high. In the three big multilateral trade agreements since 1970 that have focused on manufactured goods -- the Tokyo and Uruguay Rounds of the General Agreement on Tariffs and Trade, and the Information Technology Agreement of the World Trade Organization (WTO) -- U.S. administrations tried to satisfy both groups, and largely succeeded.

These accords, combined with free trade agreements with Canada, Mexico, Israel, and Jordan, and four duty-free programs for developing countries, have brought overall U.S. tariffs to a historic low. In July, Robert Zoellick, the U.S. trade representative, told the Bundestag Forum that "the average trade-weighted tariff for the United States is now under 2 percent."

In making this point, Zoellick was not just accurate but modest. Last year, the U.S. Customs Service collected $18.6 billion in tariff revenue on $1.1 trillion in goods imports -- meaning the effective U.S. tariff rate is 1.6 percent. But the low overall average masks something more troubling.

Tariffs on industrial imports are not just low but extremely low. For expensive consumer goods such as cars, appliances, and televisions, rates are also generally low and are further reduced in practice by trade agreements with Mexico and Canada. But for light consumer goods, the story is different. Tariffs on these products (with a few exceptions, such as toys and furniture) remain at levels other industries last saw in the 1960s and 1970s: for instance, 8.7 percent for cutlery and tableware, 13.8 percent for suitcases and handbags, 10 percent for bicycles, and 11.4 percent for shoes and clothes, the largest category of consumer imports for the United States.

In effect, the United States now has two tariff systems. One, for low-tech consumer goods, has an average rate of 10.5 percent. The other, for everything else, has an average rate of 0.8 percent. As a result, most tariff revenue now comes from a very small number of goods. Shoes and clothes in particular, as Table 1 shows, make up less than 7 percent of imports but bring in nearly half of all tariff revenues.

Finally, tariffs also vary from one consumer good to the next. Most notably, they are much higher on cheap goods than on luxuries. This disparity occurs because elite firms, selling image and brand name, find small price advantages relatively unimportant. They have not pressed the government to keep tariffs high, and tariffs on luxury goods such as silk lingerie, silver-handled cutlery, leaded-glass beer mugs, and snakeskin handbags are now very low. But makers of nylon lingerie, stainless steel cutlery, cheap water glasses, and plastic purses benefit by adding a few percentage points to their competitors’ prices. So on the cheapest goods, as Table 2 shows, tariffs are even higher than overall averages for consumer goods suggest.


So the structure of the tariff system is simple: tariffs are low overall, high on consumer goods, and especially high on cheap goods. How does it work in real life? A convenient place to begin is with the justification for the tariff system: preservation of jobs in light manufacturing.

Here, the system seems ineffective. Employment in high-tariff industries now accounts for only about three percent of U.S. manufacturing jobs. It has fallen by half since 1990, and the plunge is fastest in some of the most protected industries. In 1992, for example, 20,000 Americans worked making women’s shoes. No shoe tariffs have been cut since the 1970s, but since 1992 employment in women’s shoes has fallen by 90 percent, to only 2,000 workers. Likewise, the number of workers making children’s clothes is down from 44,000 to fewer than 7,000. For the manufacture of goods such as watches, bicycles, and drinking glasses, job totals are even lower.

In such fields, tariffs are evidently not preserving jobs. Instead they are operating more like large wholesale taxes. And although their effects are minor for domestic producers, they are considerable for people who buy and make such goods: poor families in the United States, and workers and businesses in poor countries.

One type of shoe provides an instructive example. Cheap sneakers valued at $3 or less per pair carry tariffs of 48 percent (a rate, incidentally, far above that of any product on the administration’s steel list). Virtually none of these shoes is made in the United States. Last year, the United States imported 16 million pairs of these sneakers, at a total cost of $35 million. Thus the average price at the border was $2.20 per pair. The Treasury Department then collected $17 million in tariffs, adding another $1.06 per pair to the buyers’ cost. The extra dollar and change is then magnified by retail markups of around 40 percent and state sales taxes of about 5 percent to raise the final consumer price of the sneakers from about $3.25 (without tariffs) to $4.80 per pair (with tariffs).

Such tariff-based overpricing exists, though at less dramatic levels, in store aisles stocking baby clothes, T-shirts, silverware, and other typical family products. Because it is most pronounced on the cheapest shoes and clothes, its effect falls most heavily on single-parent families.

Incomes for these families are very low: at an average of about $25,100 per year, they are about 40 percent of a typical two-parent family’s income. But single-parent families face shoe and clothing bills nearly as high as those of wealthier families. In total, the average single-parent family spends nearly $2,000 a year on clothes and shoes. Depending on the mix of purchases, as much as $400 of this total may simply be price inflation due to tariffs. Budgets for other tariffed goods, though smaller, are still a larger expense relative to income for poor families than for rich families. And so single-parent families lose much more of their income to tariffs than do other families.


Beyond U.S. shores, the tariff system operates in a similar fashion. It hits countries that specialize in the cheapest goods, in particular very poor countries in Asia, much harder than others.

Average tariffs on European exports to the United States -- primarily cars, power equipment, computers, and chemicals -- now barely exceed one percent. Developing countries such as Malaysia, which specialize in information-technology products, get rates just as low. So do natural-resource exporters such as Saudi Arabia and Nigeria. Middle-income exporters that ship a broader variety of goods, such as China, Thailand, and Brazil, face rates typically

between two percent and four percent -- above average but still not exorbitant. The least-developed Asian countries, however, take it on the chin.

For Bangladesh, Cambodia, Nepal, Mongolia, and a few others, clothes make up 90 percent of all exports to the United States. So they face average tariff rates of 14.6 percent -- nearly 10 times the world average, and 15 times the rate for wealthy Western countries. Translated into real dollars, the disparities can be remarkable.

As Table 3 notes, the U.S. now collects more tariff revenue from Bangladeshi goods than from French goods, even though Bangladesh exports $2 billion in goods a year to the United States and France $30 billion. Cambodia’s exports to the United States total $900 million and Singapore’s usually reach $15-$20 billion -- but the U.S. government collects nearly twice as much revenue from Cambodian goods as from Singaporean goods. And struggling Nepal faces tariff rates on its skirts, scarves, and suits fully 60 times higher than those applied to Ireland’s chemicals, pacemakers, and silicon chips.

In fairness, such disparities are not universal, and some attempts to ease tariff burdens have been successful. The African Growth and Opportunity Act, set up under the Clinton administration, is the best example. Since it was enacted two years ago, African clothing exports to the United States have nearly doubled, but tariff collection on African goods has dropped by more than half. The results of the Caribbean Basin Initiative are less dramatic, but Haiti, Honduras, and several other beneficiaries also face fairly low rates.

The U.S. system is not uniquely bad. Europe’s tariffs on clothes and shoes -- not to mention food -- are also high, and European Union tariff collection is far less transparent and accessible than America’s. Some developing countries would also do well to look in the mirror.

Recent World Bank studies, in fact, indicate that the poorest countries would gain most from tariff reform in large developing economies. India is a case in point: as dispiriting as U.S. treatment of Nepali goods may be, Nepal still exports more each year to the United States than to its giant southern neighbor, where clothing imports are often simply banned and tariffs on other goods regularly reach 40 percent.


The fact that others have bad policies is no excuse for Americans to adopt them too. U.S. trade policies should be defended on their own merits. And a system that hits the poor harder than anyone else is quite hard to defend.

Tariffs, especially on clothes, have tenacious advocates. But in fact, this system that protects so few and hurts so many can be fixed without much dislocation. The International Trade Commission’s July 2002 study finds that employment in high-tariff industries is already so low that eliminating all U.S. trade barriers would mean a net gain of about 35,000 jobs rather than a loss. And so, if the problems with the system were better understood, the solutions might be fairly simple.

One option, of course, would be simply to scrap tariffs on consumer goods for domestic policy reasons. Considered as a form of taxation, they are offensive to the principles of both parties -- Republicans always being enthusiastic for tax cuts, and Democrats being opposed to regressive taxation. A $10 billion annual tax cut would be substantial, but it seems moderate when compared to recent tax legislation.

Alternatively, with U.S. trade credibility needing a boost after the steel and farm bill decisions (and since the tariff policies of U.S. trading partners also need change), the administration could suggest a worldwide goal of zero tariffs on clothes, shoes, and other consumer goods. This would recapture whatever moral high ground on trade was lost in the past year and also create a powerful incentive for developing nations to seek a successful conclusion to the WTO negotiations begun at Doha last year.

But one point seems quite clear. A system that makes maids pay higher rates than corporate vice presidents, and hits Cambodians 15 times harder than Germans, is an ethical scandal and a problem far bigger than any temporary steel policy. The facts are plain, and reform is only a question of political will.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Edward Gresser is Director of the Project on Trade and Global Markets at the Progressive Policy Institute.
  • More By Edward Gresser