Does Trump Really Want to Cook the Books on American Trade?

The Wall Street Journal ran a juicy story saying so. It’s probably not true.

U.S. President Donald Trump speaks during a listening session on domestic and international human trafficking in the Roosevelt Room of the White House on February 23, 2017 in Washington, DC.
To explain whether he’s cooking the books, first we’ll need to talk about re-exports. Above, President Donald Trump in the the White House on Thursday in Washington, D.C.

Oliver Douliery/Pool-Getty Images

Over the weekend, the Wall Street Journal published a small, wonky, yet deeply disturbing scoop that seemed to confirm some of the worst fears about how Donald Trump’s White House might abuse government data. The administration was debating whether to change the way it measured U.S. trade figures, the paper reported. The proposed tweak made no economic sense, but it would conveniently exaggerate our trade deficits with countries like Mexico on paper, making it easier for the president to argue for drastic protectionist measures, like tariffs.

Given President Trump’s exhaustively documented aversion to verifiable facts, many policy experts have reasonably worried that his administration might try to fiddle with official government statistics for political purposes. The WSJ’s story suggested that, just one month into his term, the process had already begun.

But there may be less to this miniscandal than it seems. An administration source tells me that the Journal’s story contained some crucial errors. And while most readers might not be inclined to take Team Trump’s word for it, outside experts I spoke with also suggested that the Journal was simply mischaracterizing a longstanding debate about how to most accurately track trade data.

The controversy centers on what are known as “re-exports”—goods that land on U.S. soil only to be shipped somewhere else. Imagine that Walmart buys a bulk order of flat-screen TVs from a supplier in China and sends them to a distribution center in the United States. From there, it moves them to warehouses in Canada and Mexico to supply its international stores. That’s a re-export. The goods are never altered. They don’t become any more valuable during their stay here. They just get loaded onto trucks or planes and sent on their way.

Re-exports don’t really affect the American economy much and certainly don’t support manufacturing jobs. But they do get included in the monthly trade statistics that the Census Bureau reports every month, and the transactions make up a growing share of our total exports—15.4 percent last year, up from 4.2 percent in 1997, according to figures compiled by the Economic Policy Institute, a liberal think tank.

If you’re only interested in America’s global trade deficit, the presence of re-exports in our data isn’t really a problem. The goods get counted as imports on their way here. They get counted as exports as they’re sent off. Everything cancels out.

When it comes to America’s trade deficits with specific countries, however, re-exports make a difference. Think about those hypothetical TVs again. When Walmart orders them to the States from China, they’re counted as imports, even though U.S. consumers aren’t really going to buy them. When Walmart then ships them to Mexico, they’re counted as exports, even though U.S. workers didn’t make them. As a result, our trade gap looks more severe than it should with China and less severe than it should with Mexico. Re-exports distort the picture. The U.S. International Trade Commission publishes its own version of the country’s trade data that tries to correct for that problem by filtering out goods destined for re-export. But those figures tend to exaggerate our trade deficits, because while it’s possible to exclude all those Vizio TVs from our export figures, it’s not possible, for technical and administrative reasons, to entirely exclude them from our import figures.

As a result, the U.S. essentially has two imperfect sets of bilateral trade data. The Census figures that are most frequently cited in the media show that the United States had a $63 billion deficit with Mexico in 2016. According to the International Trade Commission’s figures, it was $115.4 billion. The truth is probably somewhere in between.

This brings us to the WSJ. Citing “people involved in the discussions,” the paper reported on Sunday that the Trump administration was “considering changing the way it calculates U.S. trade deficits.” How so? It wants to stop counting goods that are eventually re-exported as exports, period. At the same time, it also wants to keep counting them as imports when they arrive here, according to the paper. “Economists say that approach would inflate trade deficit numbers,” the WSJ explained, dryly, which in turn might “give the Trump administration ammunition in arguing that trade deals need to be renegotiated.”

Civil servants were reportedly displeased.

Career government employees objected last week when they were asked to prepare data using the new methodology, according to the people familiar with the discussions. These employees at the U.S. Trade Representative’s office complied with the instructions, but included their views as to why they believe the new calculation wasn’t accurate.

According to the WSJ’s reporting, it was unclear whether the administration wanted to change how official trade data were tabulated or if it simply wanted the new numbers to use as talking points while lobbying Congress. Either way, the reaction was unkind. In their morning newsletter at Politico, Ben White and Mary Lee accused Trump’s team of trying to turn the trade deficit into “an artificially bloated mirage with no basis in reality.” (They rounded up a bunch of economists and analysts to trash the idea, as well). The WSJ editorial board piled on too, accusing Trump’s staff of “trying to deceive Congress and the public with bogus data.”

But were they?

On Wednesday, an administration official with direct knowledge of the issue told me the Journal had gotten the story wrong. He said that Trump appointees had called a meeting with Commerce Department staffers to discuss whether it was possible to make U.S. bilateral trade data more precise by cutting goods that are re-exported from both the country’s export and import figures entirely. Essentially, they wanted to see if it was possible to make a more accurate version of the ITC’s data that would supplement, not replace, what Commerce already publishes. “We were talking about both sides of the equation, both imports and exports,” the official, who requested anonymity due to political sensitivities, told me. “We are looking for the most accurate data. And we understand that if we only did one side of the equation we would not have an accurate picture.”

When I asked why the administration hadn’t corrected the WSJ’s reporters when they called for comment, the official suggested they had not asked about the discussions at the Commerce Department in sufficient detail. A spokesman for the Journal told me, “We are confident in the information provided to us by our sources and stand by our original reporting.”

To be clear, it is entirely possible that the Trump team is simply trying to walk back a bad idea after being embarrassed in the pages of the nation’s leading business paper. Still, say what you will about Trump’s personal crimes against math and facts, but the administration’s story does make much more sense than what the Journal reported. After all, the “bogus data” the Journal describes sounds almost exactly like the ITC numbers that already exist and that the paper’s editorial writers have long mocked liberal trade deal critics for citing. (In fact, the follow-up editorial even accused Trump of borrowing his “trick” from the “political left.”) Why would the administration ask career staffers to cook up those figures from scratch when they’re readily available on a federal agency’s website and regularly touted by trade skeptics in Congress?

Lori Wallach, the director of Public Citizen’s Global Trade Watch, told me she doubted the Journal’s account and that trade-deal critics have long wanted a more accurate version of the ITC’s data in order to get a precise picture of what our bilateral balances look like. She added that the U.S. Trade Representative has traditionally resisted that idea and preferred citing the Census numbers, which paint deals like the North American Free Trade Agreement in a kinder light. “If the data were done right, it would show we have a much bigger NAFTA deficit than we currently do, but simultaneously it would show a smaller China deficit,” she told me. “It would be a grand irony if the Trump people”—who aren’t exactly known for their statistical rigor—“actually make that small change to the data so we can know what the heck is really going on.”

The Economic Policy Institute’s Robert Scott was similarly skeptical of the WSJ’s scoop. But he told me there were other reasons to be worried about the Trump team’s focus on re-exports. Namely, the administration seems geared toward making a political case against our trade pacts with Mexico, when it should be focusing on other countries, like China, that are doing more to destabilize global trade flows by building up massive surpluses. “To some extent it’s a big distraction, and it makes me worry about the priorities of the Trump administration,” Scott said.

It’s also debatable whether bilateral trade balances are something policymakers should be watching at all. The U.S. trade deficit is caused by a variety of factors that have little to do with the details of our individual trade deals, including the fact that Americans tend to spend more than they save. That means reducing imports from Mexico might just lead us to buy more from China or the Philippines.

But saying that Trump’s team may be unhealthily focused on bilateral trade figures isn’t quite as sexy as suggesting the administration is looking to cook the books.