Steel Quotas: A Rigged Lottery
by Gary Clyde Hufbauer, Peterson Institute for International Economics
and Erika Wada, Peterson Institute for International Economics
Gary Clyde Hufbauer is Reginald Jones senior fellow at the Institute for International Economics. Erika Wada is a Ph.D. candidate at Michigan State University and research assistant at the Institute for International Economics.
© Institute for International Economics. All rights reserved.
One of the most bizarre bills making its way through Congress is the Visclosky steel quota bill (H.R. 975). H.R. 975 passed the House on 17 March, with 289 yeas and 141 nays. It is scheduled for a Senate vote on 22 June. If enacted, H.R. 975 would limit steel imports in a heavy-handed way that scarcely benefits workers, but enriches a few lucky firms, while violating international trade rules and draining the pockets of American households. If truth-in-labeling applied to legislation, H.R. 975 would be named the Steel Lottery Act of 1999. The annual cost to American households for each steel job saved would exceed $800,000. But steel workers would receive less than 20 percent of this huge sum; lucky firms would collect more than 80 percent of the jackpot.
One curious feature of H.R. 975 is that it would augment an array of import relief measures already obtained by the steel industry. Beginning in mid-1998, the industry mounted antidumping and countervailing duty cases against steel imported from Japan, Brazil, Russia, and several other countries. In response, the Commerce Department announced penalty duties ranging up to 67 percent on Japan, 86 percent on Brazil, and 200 percent on Russia; confronted with these duties, a few countries have “voluntarily” agreed to limit their exports. More restraints of this sort are still working their way through the system.
Even more curious, the steel industry has not filed a petition for “escape clause” relief. Like the antidumping and countervailing duty penalties, escape clause relief is consistent with WTO rules. Unlike penalty duties, however, escape clause relief applies to all foreign suppliers, it demands a showing of serious injury, it is time-limited (ordinarily 4 years), and it requires the industry to make meaningful plans for adjusting to competition in the world market. These conditions may explain the industry's reticence to petition the International Trade Commission for escape clause relief.
In any event, not satisfied with existing and potential WTO-consistent remedies, integrated steel firms and their unions have made H.R. 975 a top legislative priority. The announced goals of H.R. 975 are to save steel jobs and bring prosperity to depressed mill towns. The bill requires the Commerce Department to restrict US monthly imports of steel products for the next three years to the monthly average attained during the 36-month period prior to July 1997—about 2.345 million short tons per month. If enacted, steel quotas will do little to achieve their announced goals. But like a rigged lottery, steel quotas will drain small amounts of money from many households and enrich a very few firms.
The Institute for International Economics has developed a “plain vanilla” model to estimate in a rough and ready way the costs imposed by import quotas on households, the jobs saved in the protected industry, and the windfall riches delivered to favored firms.
Protection cannot create jobs in the economy as a whole. Otherwise every country could protect itself into prosperity—a failed recipe that contributed to the Great Depression of the 1930s. But a country can save jobs in a single industry by trade protection. And it can enrich lucky quota holders and successful firms at the expense of customers. In the case of steel, that means at the expense of anyone who buys an automobile, a refrigerator, or pays taxes to construct a bridge—in other words, nearly every American household.
Applying the Institute's model to the provisions of the H.R. 975, we can ask how the contemplated steel quotas would have changed the economic landscape in the 36 months prior to July 1997—the base period chosen by Representative Peter Visclosky. His bill would have limited imports in 19 of the 36 months prior to July 1997. In these 19 months, average steel imports exceeded the average for the entire 36-month base period: 2.669 million short tons versus 2.345 million short tons. If H.R. 975 had been US law at that time, it would have lopped off the average difference of 0.324 million short tons of “excessive imports” per month in these 19 months.
Our plain vanilla model indicates that a forced drop in imports of 0.324 million tons a month would have boosted import prices in those months by about $29 a ton (this amount is mainly collected by lucky quota holders). It would also have boosted domestic prices by about $6 a ton (an amount mainly collected by successful steel firms). These price hikes would have extracted an annual average of $1.5 billion (about $125 million per month) from the pockets of American households, about $14 per household annually. Most of the jackpot money would have gone to lucky steel importers (windfall gains of $800 million) and efficient US steel firms (windfall gains of at least $400 million).
Steel workers would have been fortunate to receive $200 million in the form of pay packets for jobs saved and higher wages for workers already employed. Deadweight efficiency losses to the US economy would have run about $60 million annually.
How many steel jobs would have been saved? Our model says an annual average of about 1,700 steel jobs. Simple division reveals that the cost paid by American households for each steel job saved would have exceeded $800,000 per year. Of course nearly all of the $800,000 would go to winners of the rigged lottery, not to workers. Even allowing for a wide margin of error, this is an astonishing figure. Yet it is within the range of other estimates the Institute has made: for example, American households pay $400,000 annually for every job saved by high tariffs on ceramic tiles, $500,000 annually for every job saved by dairy quotas, and $900,000 for every job saved by tariffs on luggage. It turns out that trade barriers are a very costly way of saving jobs.
The Visclosky bill would not shut out steel imports in each and every month. Because of fluctuating demand in the US market, imports are often lower than the base period benchmark established in the bill. For example, in February 1999, imports were 2.168 million tons by comparison with the base period benchmark of 2.345 million tons. But in March 1999, imports rose to 2.845 million tons. In other words, the bill would have permitted all of February imports, but shut out 500,000 tons of March imports. This kind of fluctuation highlights the disruption caused by monthly quotas: by shutting out steel imports in some months and permiting imports in others, they add to the logistical burdens on US users and foreign suppliers. The numerous House co-sponsors of the Visclosky bill possibly regarded “supply harassment” as an extra virtue of their legislative handiwork.
It is worth mentioning that the costs imposed by a monthly quota would soar in times of high US demand. For example, in the third quarter of 1998, when imports averaged 3.922 million tons per month, the annualized cost to American households would have exceeded $6.6 billion. To be sure, 6,700 steel jobs would have been saved during those three months, but at a cost approaching $1 million per steel job saved.
The dynamic US economy, which at full employment creates 2 to 3 million net new jobs annually, also lays off about 1.4 million workers annually, as plants close and firms downsize. Even when the economy is enjoying full employment, about a quarter of laid-off workers remain unemployed for several years. That fact alone provides a compelling argument for national training and relocation programs. But it is not an argument for a rigged lottery that would take $1.5 billion per year from American households, confer $1.2 billion of windfall profits on a few firms, yet save the jobs of only 1,700 workers at an annual cost exceeding $800,000 per worker.