Last month, President Donald Trump promulgated a memorandum on “Steel Imports and Threats to National Security,” directing Commerce Secretary Wilbur Ross to consider whether “steel is being imported into the United States in such quantities or under such circumstances as to threaten to impair the national security.” Prices are too low, he wrote, because the world is awash in excess capacity. The memo notes that Ross has already initiated an investigation under Section 232 of the Trade Expansion Act of 1962 (19 U.S.C. 1862). If he finds a problem, he may recommend “steps that should be taken to adjust steel imports so that they will not threaten to impair the national security.” Governments mostly “adjust” trade with inefficient and unexpected results, so before taking “steps,” we should ask three questions. When have steel imports been problematic to national security in the past? Is the situation presently getting worse? And what circumstances might exacerbate the presumed problem in the future?
Past. Consider the most recent stressing case: the sudden need for thousands of new armored vehicles during the Iraq War. Armoring trucks and Humvees, and building Mine-Resistant Ambush-Protected Vehicles (MRAPs), required much more armor-grade steel than the US Army and Marine Corps might buy in any given year. In my recent dissertation on the MRAP program (p. 169), I observed that “no bottleneck was as threatening as that of steel plate” in this production. Each of the thousands of MRAPs built required four tons of steel plate that industry just wasn’t producing. Back in 2006,
Only two domestic suppliers were certified by the Defense Department to supply the 3/8-inch armor plate used in MRAP construction: International Steel Group and Oregon Steel Mills.  Only the former was actually in operation at the start of the war. Its production was 35,000 tons annually in 2004, and had been only 6,500 tons annually in 2003. The DoD would suddenly be demanding 21,000 monthly.  Interestingly, International was owned by Arcelor Mittal, the world’s largest steelmaker; while headquartered in the Netherlands, its chairman and largest shareholder was Indian steel magnate Lakshmi Mittal. More interestingly, Oregon Steel Mills was owned by the Evraz Group; while its shares are traded on the London Stock Exchange, the company itself was headquartered in Luxembourg City and Moscow, and was the largest steel producer in Russia. [p. 169]
Getting attention from American industry for this sort of build-up should have been difficult. Under the Defense Production Act of 1950—a statute which grants the executive breathtaking economic authority in an emergency—the president may demand preferential treatment from any supplier based in the United States. The most comprehensive authority is delegated only to the Secretary of Defense, who can assign any program a “DX” rating, sending its military orders to the head of the line. The most recent revision of the the DX list (1 March 2016) includes just nine programs, mostly related to nuclear weapons: Air Force One, the presidential helicopters, B-2 bombers, Minuteman and Trident nuclear missiles, the ballistic missile defense system, the SBIRS launch-warning satellites, the Program 390 reconnaissance satellites, and the mysterious Naval Program 341. In June 2007, Secretary Gates accorded the MRAP a DX rating, indicating that he really wanted to win that war. So did some unusual suspects:
Those foreign-owned firms absolutely supported the war effort. Mittal USA spokesman David Allen told National Defense that “DX ratings are nothing new. We’ve seen them on-and-off since 1980. We would respond to any DX rating as we have in the past.”  More so, SvenskaStål AB (SSAB), a Swedish specialty steel manufacturer well-known for armor plate, had already supplied two MRAP producers, including Force Protection. SSAB had other priorities too, including supplying steel for the booming business of erecting cellular telephone towers in China. Potential supply shortages through its mills in Sweden were eventually addressed by intergovernmental discussions. Force Dynamics diversified its sources of steel for components other-than-armor in 2007, adding Canada’s Algoma Steel to Mittal as a second supplier on the Cougar program.  That year, though, Algoma was purchased by India’s Essar Group, and retitled Essar Steel Algoma. In a globalizing economy, there was simply no wholly American, or even North American, solution to the industrial problem. [pp. 169–170]
I should be clear that long-standing American defense contractors were responding. They needed steel plate to up-armor all those soft-skinned trucks, but they sometimes endured a roundabout process to get it:
In December 2004, the Army’s deputy chief of staff for programs directed that all wheeled vehicles would be armored before “crossing berm” from Kuwait.  Each of the [original equipment manufacturers] for the Army and Marine Corps’ trucks received a share of the work… Oshkosh Truck of Wisconsin received a sole-source contract to up-armor the Marine Corps’ Medium Tactical Vehicle Replacements (MTVRs). Oshkosh received similar contracts to armor-up its other large military trucks, too—the Army’s Family of Heavy Tactical Vehicles (FHTVs), and the Marine Corps’ Logistic Vehicle System Replacement (LVSR).  The company installed armor kits built by Plasan in Israel with American steel.  Using American steel was mandated by the Berry Amendment (10 U.S.C. 2533a), which requires that federal funds expended for certain categories of products be spent only on domestically manufactured items, unless there are no reasonable substitutes. Thus, American steel was shipped to Israel to be made into Israeli armor plate to be shipped to the United States. [pp. 81–82]
There is nothing economically or militarily sensible about that last sentence. The primary losers were the troops in the field, who had to wait for armor so that American industrialists and steel workers could get paid. In this case, the Berry Amendment didn’t further national security; it hindered it. Late in the Bush Administration, the industrial policy and acquisition staffs were working to secure additional sources. But as a former Pentagon official told me the other day, they had to turn to Russian, Indian, Canadian, Swedish, and Israeli-owned firms because the American-owned ones “told us to pound sand.” And why not? They were making money by administrative fiat, but the business just wasn’t that much of their order books. They could have expanded production, and then been left with stranded assets after the fighting ended.
That’s because in the war effort, the question was not steel, but specialty steel, and not much of it. The less than two hundred thousand tons that went into all the armor for the troops represented about two-tenths of one percent of American steel production during a single typical year in the 2000s. Yet at the time, outside shipbuilding, armoring trucks was the largest use of steel in military production anywhere. The MRAPs also constituted the single largest armored vehicle program since the Second World War, so if their demands for steel failed to confound the supply chain, then it’s safe to say that the US hasn’t had an overwhelming problem at any point since the 1940s.
Present. President Trump’s interest in the steel industry became plain by last June, when Candidate Trump gave a speech at an aluminum recycling plant in Monessen, Pennsylvania. At the time, KDKA Pittsburgh noted that the site employed 35 people, down from the several hundred it employed when it was the Monessen Works of Wheeling-Pittsburgh Steel. (There was no word about which would be better for America in the long run: making new steel, or recycling old aluminum.) The next day, while meeting in Ottawa with Prime Minister Trudeau, then-President Obama asserted that “the [American] steel industry is producing as much steel in the United States as it ever was. It’s just (that) it needs one-tenth of the workers that it used to.” Politifact checked; Obama wasn’t quite right, but for the purposes of the analysis, he wasn't too far off.
In 2015, the most recent year of full information collected by the Minerals Resources Program of the US Geological Survey, domestic shipments of steel products amounted to 78.5 million tons. That figure was down 22 percent from its peak of 101 million tons in 2004. In 2015, domestic production of raw steel was 78.8 million tons; that had peaked in 1973 at 137 million tons. We might further discount the decline for economic growth, for the United States now has more than twice its population back then, with still vigorous demand for cars and bridges and buildings. However, at 99.8 million tons in 2015, steel consumption is also down, from its all-time peak of 120 million tons in 2000. New materials have supplanted some of steel’s uses.
According to the American Iron and Steel Institute, direct employment in the domestic steel industry is about 142,000 people. That’s down 78 percent from its peak of 650,000 in 1953. Obama’s assertion was actually an overestimate, but he alluded to the reason: a roughly five-fold increase in labor productivity just since 1980. Along the way, that made for a short-term loss for 508,000 workers, but a lasting benefit for Americans as a whole. In the long run, all that labor became available to do other productive things. Ceteris paribus, this factor-productivity improvement has also improved American security. Should wartime requirements someday mandate expanding steel production, staffing up would be much easier. Each worker recruited would need to be more skilled, but industry would need to find only one—not five.
As chronicled by the Bureau of Labor Statistics, steel prices are sharply down, to be sure. However, that change must be considered in the context of the large increase in steel prices after 2003, their crash in 2008, and their substantial recovery by 2010. In recent years, perhaps a few Chinese red chip companies have been over-investing in new furnaces and mills (see below). But how low steel prices affect national security overall is less clear, for they certainly make the US Navy’s ships cheaper (see further below).
Of late, however, US imports have increased, in both absolute terms and as a percentage of consumption. In 2015, about 35 percent of the steel products consumed in the US were imported. Then again, in only one year since 1993 has that figure been less than 20 percent. It is also notable that in every year since 2010, over 10 percent of production of American steel goods has been exported. In 2015, exports totaled about 9 million tons, a figure never seen before 2007. American exports of raw steel are much less, but this has also been the case since the mid-1990s. All this suggests that certain sectors of the American steel industry are doing just fine for now, relative to foreign producers. That the entire worldwide industry is having a hard time is another matter, and possibly not a security matter.
Future. The question is not endemic to the US. In April 2016, James Wilson, Michael Pooler and Kiran Stacey of the Financial Times considered the question “is UK steel really a strategic industry?” Steel production there about tenth of a percent of gross domestic output. Britain has only three large steel plants, all owned by the Indian firm the Tata Group. The future of those factories is in doubt. Having purchased all of British Steel in 2007, Tata now wants to sell; the Guardian cited “cheap imports of Chinese steel, high energy costs, and weak demand.” Indeed, about half of the 1.6 billion tons of steel produced annually worldwide come from China, and that national market share has held for about a decade.
But does this matter? By tonnage, almost half of the world’s merchant shipping is built in China too, but certainly none of NATO’s warships are. The Royal Navy’s aircraft carriers are built from British steel, but its submarines get their from France, and the British Army’s new Ajax armored vehicles get theirs from Sweden. The US produces far more steel than its military needs, but even if the UK can’t supply its own domestically, it can easily buy the steel it needs from reliable and nearby sources.
In the US, industry likes to talk up its game, alluding to a frightening future without domestic steel production. In its somewhat dated flyer “Backbone Industry in National Defense,” the American Iron and Steel Institute reminds us that “the ten [aircraft] carriers constructed in recent years consumed 500,000 tons of steel.” If recent years means the last half century, then that’s about right. The US Navy buys about one super-carrier every five years, so annual requirements are about 10,000 tons. Add in the rest of American naval shipbuilding—roughly two destroyers, two submarines, two frigates, an assault ship, some cutters, and a few cargo ships each year—and the annual total does not exceed 100,000 tons. Important speciality steels go into aircraft production; the tonnages are just trivial in this context. Add in the ground forces’ needs for vehicles, and in sum American military requirements still draw considerably less than one percent of domestic steel output.
But should Defense and Commerce worry in this regard about the Chinese? As noted, they have half the world’s steel output, almost half the world’s shipbuilding, and a large and growing navy. This sounds roughly like the problem that the Japanese Empire faced in choosing to attack in 1941—just in reverse for the United States. Today, the US isn’t in the middle of a massive war that consumes steel for vast quantities of tanks, ships, and armaments. But could it be? And then what would the Army and Navy do?
For context, note also that the American steel industry is producing more steel product today than it was during the Second World War. The wartime peak was just 60.5 million tons in 1943. At the end of that war, the US was producing about 70 percent of all raw steel worldwide, but the difference was in the denominator: German and Japanese steel had been bombed out of existence. Today, the US is producing roughly the same amount of raw steel as it did then; the excess was exported, as American foundries were feeding factories and shipyards in Canada and Britain. All of this was enough for the huge armadas that defeated the Axis Powers, and the US actually has more capacity today. If anything, it’s shipbuilding capacity that the Navy lacks for an emergency. If there is to be a future bottleneck in a global war, it will be found in skilled shipbuilders and graving docks, far before we’re searching for more steel plate.
Summary. So how should we assess the effect of steel imports on national security? In the short run, anticipate that foreign suppliers will respond—and maybe faster than Americans—just as they did during the Iraq War. In that regard, steel imports haven't impaired the national security. On the contrary, they have improved it. In the future, if there are any difficulties with existing overseas suppliers, those “intergovernmental discussions” can be helpful. In the long run, relative to its current and future military needs, the US has vast excess capacity for steel production. If there is an issue, as evidenced by the MRAP experience, it is that the US may lack the specialized capacity for all its military needs. As the commerce secretary investigates, he might profitably direct efforts into the details; defense-industrial problems are often not as simple as the headlines might suggest.
James Hasik is a senior fellow at the Brent Scowcroft Center on International Security.