| 2019-2025 | 94 million tons |
| 2000-2017 | 104 million tons |
** U.S. Geological Survey, Mineral Commodity Summaries. Excludes 2018 as half-tariffed, half not; that year’s consumption was 101 million tons. An alternative count from the World Steel Association has a slightly smaller disparity: 91.4 million tons on average from 2019-2025, 99.4 million tons from 2000-2017.
In the world of “industrial strategy,” is “subsidizing abundance” or “manufacturing scarcity” the better approach? A case study of Trump administration steel policy, with the Biden-era semiconductor CHIPS program as the “abundance” alternative –
Background: In 2018, the first-term Trump administration imposed “national security” tariffs of 25% on most imported steel and 7.5% on most aluminum. (Before that, steel was, in theory, mostly duty-free, but in practice frequently subject to anti-dumping penalties. See below for detail.) In February 2025, the second-term version raised the aluminum rate to 25% and extended steel and aluminum tariffs alike to Canadian and Mexican metal in February 2025; then, four months later, it hiked the rate for both to 50% and added “semi-finished” copper products (but not raw/refined copper metal) as well.
What’s happened since? Three data points and a couple of tentative conclusions:
Prices have gone up: According to the Department of Commerce’s quarterly “Steel Executive Summary,” steel in the U.S. was already pricey two years ago. In January 2025 – just before the inauguration — American metal-buyers paid an average of $758 per ton for steel, as against a worldwide average of $456 per ton. (Country detail: the European average was $587/ton, and the Chinese $386/ton.) Since then, world steel prices have been mostly stable, while the U.S. price has risen impressively — hitting $950/ton by mid-summer 2025, then stabilizing for a while, then jumping again to $1,200 this summer, more than double the $490 worldwide average.
| Date | U.S. Average | World Average |
| June 2026 | $1,208/ton | $490/ton |
| January 2026 | $978/ton | $455/ton |
| June 2025 | $960/ton | $440/ton |
| January 2025 | $758/ton | $456/ton |
* Commerce Department Steel Executive Summary, April-June 2026
Aluminum trends are similar: American buyers now pay roughly $4,500 per ton, about $1,000 above the $3,500 worldwide average. Copper is more expensive too, but the exemption of “cathodes, bars, and ingots” means the price gap isn’t very high.
Output about the same: The U.S. Geological Survey’s annual “Mineral Commodity Summaries” says U.S.-based companies produced 82 million tons of steel last year. This is a bit above the 81.3 million-ton “tariff-era” annual average from 2019 to 2025, and a bit below the 84 million-ton “pre-tariff” average from 2010 to 2017.*
Steel use has dropped: Though output hasn’t changed much, Americans seem now to use less metal than they did a decade ago. The USGS says last year’s “apparent consumption” of steel – for buildings, cars, machinery, etc. – totaled 95 million tons. This was 5 million tons less than the 100 million tons used in 2021, and 9 million tons below the 104 million-ton average from 2000 to 2017. Or, if you use the World Steel Association’s figures instead of USGS’s, 8 million tons less than the 2000-2017 average.
What does this suggest? Basically, the Trump administration’s metals program has been one of “manufactured scarcity.” That is, it has, with some success, tried to make steel more expensive, with less diversity of supply, more market power for U.S.-based producers, and a smaller overall U.S. market. (Aluminum too; less so copper.) This strategy can make U.S.-based producers more profitable, but with predictable consequences:
(i) “Buyers pay”: The administration justifies these policies as a “national security” measure – that is, a nationwide priority for which the American public should be willing to pay a cost. The approach Biden administration’s semiconductor program five years ago — “CHIPS” — took a that approach, investing $54 billion in chip production and research through grants and tax breaks. Overall, this aimed to create more chipmaking without raising prices or cutting supply. The Trump administration’s program, by contrast, asks metals buyers alone – not the public as a whole – to pay for the ‘security’ through higher prices and fewer supply options. In practice, the “metals buyers” are U.S.-based automobile manufacturers, construction firms, federal government infrastructure projects, machinery-makers, and so on. As an example, a compact SUV contains about a ton of steel and 200 kilos of aluminum. The widened disparity between U.S. and world steel prices means a U.S.-made vehicle starts out about $500 in the hole as against a European car, and nearly $800 for a Chinese car. Aluminum disparities add about $400 more.
(ii) “Buyers respond”: Faced with this cost, the buyers have some choices: shrink a bit, shift costs onto the public by raising their own prices, or innovate so as to use less metal. Some of each may be going on. The GDP “share” of big metal-using industries — in Bureau of Economic Analysis jargon, construction, “fabricated metal products,” automotive industry, and “other transportation equipment” — has dropped, with the manufacturers’ share down from 3.1% in 2017 to 2.6% as of early 2026, and the construction firms’ from 4.4% to 4.3%. (“Primary metals” have been stable at 0.3%.) Costs have gone up a bit — inflation in construction appears to be about double the national average, and new car prices are up about $2,000. At the same time, American steel use has dropped about 10% from 2000-2017 levels.
Closing thought: The idea that primary metals production has “national security” value isn’t obviously wrong. As with semiconductors, metals are inputs for “downstream” industries of all kinds, and diverse and reliable supply is important. As to whether the U.S. metals companies genuinely need an expensive public “industrial strategy” program, opinions can vary. But either way, the administration’s program — “manufactured scarcity”, which aims to help local producers by raising prices and reducing competition — has the basic flaw of discouraging metal use and perhaps reducing productivity over the long term. If national security requires something expensive, the Biden administration’s more “abundance-based” approach to semiconductors — spend to promote production, but don’t raise prices or diminish competition — looks fairer and better.
* Excluding 2018 as “half tariffed, half not”; that year’s output was 86.6 million tons.
PPI’s four principles for response to tariffs and economic isolationism:
Metals:
The U.S. Geological Survey’s annual “Mineral Commodity Surveys” have concise two-page summaries of output, imports, exports, consumption, and more for iron and steel. Spreadsheets carry the data back to 1900.
The Commerce Department’s Steel Executive Summary (April-June 2026) has imports, exports, production, prices, etc., for steel.
The World Steel Association’s country-by-country data on steel.
Costs:
Cox Automotive tracks new car prices.
From the Federal Reserve’s “FRED” series, an inflation-in-construction-materials graph.
And “GDP by Industry” data, including GDP shares, from the Bureau of Economic Analysis.
CHIPS perspectives:
Commerce Department 2022 program summary.
Military/security perspective from the Defense Advanced Research Projects Agency
The National Institute for Standards and Technology has grant updates.
State of Arizona view.
… and TSMC Arizona status.
And some steel-tariff context:
If U.S. metals tariffs are 50% now, what were they before? Kind of a murky question, actually.
Most permanent U.S. tariffs on metals are zero or 1%, though some are a bit higher. (See Harmonized Tariff Schedule Chapters 72 and 73 for steel, 74 for copper, and 76 for aluminum.) Aluminum and copper are pretty straightforward, with actual rates generally the ones published in the HTS. Steel products, though, also carry 333 anti-dumping and countervailing duty orders,
about 40% of the 846 such orders on all products combined. So in practice, pre-2017 steel from different countries got very different treatment and was subject to rapid changes. Obama-era cases, for example, cut off about 90% of Chinese steel imports in 2015.
All this means that (a) the pre-2017 tariff rates on steel are not easy to calculate, and (b) the Trump-era tariffs really fall mainly on European, Japanese, Indian, Korean, Brazilian, British, and more recently Canadian and Mexican steel. See DoC’s anti-dumping and countervailing duty dashboard, searchable by product, date, and country.
Ed Gresser is Vice President and Director for Trade and Global Markets at PPI.
Ed returns to PPI after working for the think tank from 2001-2011. He most recently served as the Assistant U.S. Trade Representative for Trade Policy and Economics at the Office of the United States Trade Representative (USTR). In this position, he led USTR’s economic research unit from 2015-2021, and chaired the 21-agency Trade Policy Staff Committee.
Ed began his career on Capitol Hill before serving USTR as Policy Advisor to USTR Charlene Barshefsky from 1998 to 2001. He then led PPI’s Trade and Global Markets Project from 2001 to 2011. After PPI, he co-founded and directed the independent think tank ProgressiveEconomy until rejoining USTR in 2015. In 2013, the Washington International Trade Association presented him with its Lighthouse Award, awarded annually to an individual or group for significant contributions to trade policy.
Ed is the author of Freedom from Want: American Liberalism and the Global Economy (2007). He has published in a variety of journals and newspapers, and his research has been cited by leading academics and international organizations including the WTO, World Bank, and International Monetary Fund. He is a graduate of Stanford University and holds a Master’s Degree in International Affairs from Columbia Universities and a certificate from the Averell Harriman Institute for Advanced Study of the Soviet Union.
