Congress Should Retire, Not Reform, the Generalized System of Preferences



Marc L. Busch | The Hill

The United States’ Generalized System of Preferences (GSP) is set to expire at the end of the year. This trade program, which started up in 1976, grants developing countries zero-tariffs on eligible goods. U.S. Trade Representative Robert Lighthizer says he may want to reform GSP before renewing it. GSP can’t be reformed. Instead, Congress should start the process of retiring it.

The U.S. extends one of 13 GSP programs in today’s global economy. All of them aim to help poor nations boost their exports through tariff preferences. U.S. GSP, in particular, comes with political conditionality and caps annual import growth. The U.S. “suspends” recipients that fall short on labor standards or intellectual property rights, for example, and has “competitive limitations” on a recipients’ year-on-year exports. These design features rub poor nations the wrong way, but the main problem with GSP is what it does inside developing countries.

At first blush, GSP sounds like a free lunch because developing countries get tariff cuts without having to give them. But it’s not. One indication is that GSP is underused: The average utilization rate is a respectable 72 percent, but for some of the poorest nations, such as Ghana, Lesotho and Sierra Leone, it’s only 27 percent, 38 percent and 36 percent, respectively.

The reason for this underuse is that GSP’s margin of preference, in relation to the most-favored nation (MFN) tariffs under the World Trade Organization (WTO), isn’t big. It’s certainly not big enough to take on the risk of being suspended or staying below competitive limitations. Things were different in the 1970s. Most recipients didn’t belong to the General Agreement on Tariffs and Trade (GATT), so GSP’s margin of preference was sizable. Today, far more recipients are members of the WTO, such that GSP’s preference margin over MFN averages a mere 2.4 percent.

This underutilization of GSP isn’t the problem. The problem is that GSP does bad things to the domestic trade politics of recipients that belong to the WTO. 

The issue lies with the exporters. They get market access abroad regardless of whether their government liberalizes. That’s because GSP is nonreciprocal. But conditionality still looms large. That’s where the WTO comes in. Although not a WTO obligation, GSP is permitted by the WTO and has been the subject of litigation. The ruling says that conditionality has to be applied the same way across recipients in similar situations. The WTO, in other words, helps to insulate exporters from ad hoc conditionality, and thus reduces their incentive to lobby against tariffs at home. This leads the recipient to import less, hurting consumers and industries that make use of imported inputs.

So, what is there to reform? GSP works exactly as intended. It’s just that GSP happens to distort trade politics in recipients that belong to the WTO. This wasn’t entirely unanticipated. In 1968, at a conference in New Delhi that brought the idea of GSP to life, developing countries asked how the program would work with the GATT’s multilateral rules. It didn’t fit back then, and it can’t be made to fit now. 

Proponents of GSP will push back. One line of argument is that developing countries are unable to go without GSP. This is a stretch. GSP covers only about 3 percent of U.S. imports. Even India, a relatively large user of GSP, reported little change in exports to the U.S. after being suspended in 2019. India also wants a free trade deal with the U.S. That’s the future, not GSP.