What if it were true? What if US aluminum tariffs filled Washington’s coffers, created American jobs, and did not hurt the country’s downstream industries or consumers? A win-win, right? That’s the claim made in a new study by the Economic Policy Institute, remarkably titled “Aluminum Producing and Consuming Industries Have Thrived Under US Section 232 Import Measures.” If this was in fact true, it would be a game-changer, right?

But it’s not true.

Let’s start from the beginning. The study looks at President Trump’s Section 232 tariffs on aluminum. It argues these 10% import duties “had no adverse effect on downstream industries or consumers.” It gets better. The study insists that production capacity, investment in facilities, and jobs all increased as a result of the tariffs. In other words, all benefits, no costs.

But it’s not true.

The most glaring problem with the study is that it does not actually have a measure of harm to downstream end-users or consumers. It looks at whether producers of beer cans, for example, paid higher prices for aluminum because of the tariff, and passed these along to the consumer. The report says no and no. But to see why this doesn’t make much sense, consider an example.

One of the American companies that got hit with a 25% tariff when the European Union (EU) retaliated for Trump’s tariffs was Harley-Davidson. The tariff added $2,200 per bike sold in the EU. But Harley-Davidson ate the tariff, meaning it incurred the cost without passing it on to buyers. It would be ridiculous to claim the tariff didn’t hurt Harley-Davidson, but that’s exactly what the study would have you believe, because it only looks at prices.

Why would Harley-Davidson eat these tariffs? Competition, for one thing. Or inventory on hand in the EU that could more easily be “priced to market.” The study does not model any of this. Instead, it uses a simplistic tool that attributes all the action to change over time, using one control variable. To figure things out, a study of this sort has to examine the relative tariff exposure of downstream industries and employment growth across them, so as to pick up the cross-sectional variation of the effects of tariffs, not just time.

In truth, the study doesn’t believe its own finding. If it did, the conclusion wouldn’t just encourage the Biden-Harris administration to “continue” the tariffs. Rather, it would advocate raising these tariffs. Think about it: if tariffs equal more government revenue, more production and investment, more jobs, and cause no harm, then the U.S. needs more of these tariffs, right? This is tariff alchemy: win-win.

Done right, a study of this sort would also calculate the cost per job saved (if any) in the tariff-affected industry. There’s nothing new about showing that a tariff might result in a short-run uptick in employment, but typically at a cost that greatly exceeds what these jobs pay. The study says nothing about the cost per job saved. Again, all benefits, no costs.

Unbelievably, the study argues that foreign retaliation isn’t costly either. It takes a very cursory look at American spirits. Did the EU’s 25% retaliatory tariffs on bourbon, rye and American whiskey hurt exports? The study says no. What’s the evidence? The study quotes the Distilled Spirits Council of the United States about the rising popularity of American spirits, and says global trends, not EU tariffs, are to blame for any woes. But the Council could not disagree more: it finds that U.S. exports of spirits “plunged 37%” to the EU and by a whopping 53% to the United Kingdom. Is this what “no harm” looks like?

One last point. Let’s assume for the moment that tariff alchemy is real. Why use Section 232? The study taps protectionists’ favorite fiction that foreign firms are all subsidized, making it possible for them to sell below their cost and purge U.S. firms left to fend for themselves. If so, then why not use an antidumping or countervailing duty, or a safeguard? Why Section 232?

I’ll tell you why. It’s because Section 232 is the new holy grail of protectionists. A tariff under the banner of national security is much more versatile. It is not constrained by the legal requirements of antidumping or countervailing duty, or a safeguard. And once the genie is out of the bottle, and other countries use their own national security tariffs, the protectionists can call for still more U.S. retaliation.

But it’s not true.

In reality, U.S. companies need imported aluminum to stay in business. Harbor Aluminum found that only one in five tons of primary aluminum consumed in the United States is produced in the United States. In fact, a 40-year-old smelter, built in 1980, is the most recent American smelter working today. The U.S. can’t meet domestic demand without imported aluminum, a conclusion that should add to the case for the Biden administration to end Trump’s costly aluminum tariffs.

Marc L. Busch is the Karl F. Landegger Professor of International Business Diplomacy at the Walsh School of Foreign Service, Georgetown University, a nonresident Senior Fellow at the Atlantic Council, and host of the podcast TradeCraft