Author: Justin R. Pierce (Board of Governors of the Federal Reserve System)
Publisher: Federal Reserve Board Finance and Economics Discussion Series Paper No. 2011-40
Date Published: September 2011
Antidumping duties, designed to protect domestic industries from unfair foreign competition, play a critical role in international trade. Such tariffs are placed on notably low-priced imports. In the United States, these imports are considered to be offered at less than fair value — or “dumped” — when the price is lower than what is charged for comparable goods in the home market, or below an adjusted value tied to average total cost.
The duties are popular in the U.S. — in recent years, companies or unions from 16 of 19 manufacturing sectors have filed petitions seeking such protection. The Department of Commerce determines whether dumping has occurred and then the International Trade Commission decides whether the U.S. industry has been harmed. If so, a tariff equal to the percentage difference between the price in the U.S. and the price in the home market is slapped on the imported goods. The slap can really hurt. Some years ago, a 259 percent duty was placed on aluminum sulfate from Venezuela, for example, leading to a 98 percent plunge in U.S. imports from that country.
But despite the widespread appeal of antidumping penalties among U.S. companies and unions, their effect on the sale of U.S. domestic products has remained murky. Although some studies have found a correlation between antidumping protection and increases in revenue, this paper argues that those findings are misleading. Temporary gains in revenue are merely a result of short-term price hikes put in place by the protected U.S. manufacturers, the author says, while the number of units produced domestically dips. What’s more, antidumping duties could have a downside in that they enable firms to continue operating plants that are under-producing, in turn wasting resources.
“Increases in prices and markups artificially inflate the effect of antidumping duties on revenue productivity, while physical productivity actually falls,” the author writes. “Moreover, antidumping duties allow low-productivity plants to continue producing protected products, slowing the reallocation of resources from less productive to more productive uses.”
This study, which the author calls “the first micro-level evidence on the effects of antidumping duties in the United States with a dataset that includes the full population of U.S. manufacturing plants,” compares the behavior of plants that receive protection against a control group of plants in similar industries that do not. Crucially, the perceived effect of antidumping penalties on plant-level production depends on whether output is measured in revenue or in physical units, because price hikes and markups would likely be taking place at the same time as any shifts in physical productivity.
To identify the source of the revenue boom for protected firms — whether it is a jump in the quantity of goods produced or a factor of short-term markups — the author analyzed data from the U.S. Census Bureau’s Census of Manufactures for the years 1987, 1992, and 1997. This 10-year period was chosen because it was a stable era for the Census’s Standard Industrial Classification codes, with only minor changes to product-class codes. Several recent studies, including a 2010 analysis of the product-switching behavior of U.S. manufacturers, have relied on this data set, noting that the underlying dynamics of the antidumping environment have remained largely consistent.
Irrespective of size, all U.S. manufacturers are required to respond to the Census. The information contains plant-level data on the value of shipments, the number of employees and their roles, raw material usage, and the book value of capital assets, which can be used to measure productivity. The Census also provides plant-level and product-level output data in terms of revenue and physical units for each product — a significant feature because it allowed the researcher to calculate physical productivity as well as the average unit prices and price-cost markups.