(This column has been updated)
A reader asked us to explore this assertion of 85,000 lost jobs from the U.S.-South Korea Free Trade Agreement, noting that a year earlier Public Citizen asserted that an increase in the trade deficit amounted to 70,000 lost American jobs. How was this possible, he asked, when the economy is booming — and U.S. trade data show that U.S. exports increased since the Korea FTA was implemented?
The question may seem obscure but it is increasingly important in the debate over the pending Trans-Pacific Partnership, a 12-nation deal for the Pacific Rim. Public Citizen is influential on the left in the trade debate, and its methodology for calculating these Korea FTA numbers mirrors its assertion that the United States is running a trade deficit, not a surplus, with all FTA partners. Some lawmakers have adopted Public Citizen’s math, saying the administration is “cooking the books” on trade.
While Public Citizen says they are not claiming 85,000 jobs have actually been lost, that nuance has been lost on some members of Congress. “The U.S.-Korea Free Trade Agreement has led to an estimated 84 percent increase in the U.S. trade deficit with Korea, equivalent to $12.7 billion and nearly 85,000 jobs lost since its ratification just three years ago,” reads a recent news release by Rep. Tim Ryan (D-Ohio).
Trade is a complex issue, making its susceptible to political gamesmanship. Readers will recall that The Fact Checker gave Four Pinocchios to the administration’s claim that the TPP would result in 650,000 additional jobs. How does Public Citizen’s math hold up?
The Korea FTA is the second-largest such agreement, after the North America Free Trade Agreement (NAFTA). It went into effect on March 15, 2012, reducing and eliminating tariffs and non-tariff barriers to trade.
The story of Public Citizen’s estimate starts with a White House “fact sheet” from 2011, in which it asserted that ratification of the Korea deal is “estimated to support 70,000 American jobs from increased goods exports alone.” The calculation was based on an estimate by the International Trade Commission that the trade agreement would increase exports of American goods by $10 billion to $11 billion.
Regular readers know we warn about taking any these jobs claims at face value. In this case, a relatively simple formula was applied, in which every $150,000 of exported goods was deemed to have supported each job. ($10.5 billion divided by $150,000 equals 70,000.) That’s not out of line, though the White House ignored that fact that the ITC also estimated that imports would increase as well, by nearly $7 billion, potentially offsetting at least some of those supposed gains.
The 400-page ITC report is dense and complex, methodically going through just about every product that could have been affected by the free-trade deal. But the figure on exports was a “comparative static” estimate, meaning the ITC took just a moment in time in 2007 and tried to figure out the change in trade flows if all other factors were held constant. “These simulation results should not be interpreted as changes in total imports and exports, or as implying meaningful information about the balance of trade impact of the entire U.S.-Korea FTA,” the report warns.
Moreover, the estimate was for when the FTSA would be fully implemented — in 20 years (though most tariffs will be eliminated by 2022) — but using estimated data for 2008. Overall, the report said the U.S. gross domestic product would increase by $10 billion to $12 billion, for an increase of 0.1 percent — essentially a rounding error. The impact on employment was also deemed to be negligible. So there really wasn’t much to the White House’s 70,000 jobs estimate in the first place; it would have earned Pinocchios if it had come to our attention at the time.
Moreover, the numbers in that 2007 report were out of the date by the time the FTA was launched. The report notes that “U.S. merchandise exports to Korea were valued at $30.8 billion in 2006.” By 2012, it was nearly $40 billion, meaning that supposed $10 billion gain had been achieved just a few months after the FTA was launched. But even trade advocates are not claiming that.
So how does Public Citizen claim that the same methodology results in 85,000 job losses, based on the trade data from three years of the Korea FTA? It requires some fancy footwork.
Public Citizen notes that the report says that the merchandise trade deficit with Korea was $13.9 billion in 2006. Using what it says is the same methodology as the report, the group calculates that the merchandise trade deficit has increased by about $12.5 billion in the three years since the Korea FTA was implemented. (Public Citizen oddly also factors in inflation, which strikes us as an effort to manipulate the data further, since the Bureau of Labor Statistics inflation rate reflects costs throughout the economy–the price of goods could decrease.) Then Public Citizen divides that figure by $150,000 (each job supported by exports) and comes up with a loss of 85,000 jobs.
But the White House estimate was not based on a trade deficit number, but on a gain in merchandise exports.
Public Citizen claims they are trying to be very precise in their numbers, measuring from April 2012 because the FTA went into effect on March 15, 2012. But by lopping off February and March 2012, they remove two months of very high export numbers as exporters were preparing for the new trade regime. (Their data for March 2015 is also an estimate.) Measuring from the end of February 2012 through February 2015, the most recent month for which data is available, yields a gain in exports over three years.
In any case, the most appropriate way to look at exports flows would be on an annual basis, which shows a gain of about $1.1 billion from the end of 2011 to the end of 2014. That’s theoretically a gain of more than 7,000 — a far cry from a loss of 85,000. (Note: an earlier version of this article offered a calculation from the end of 2012, which shows a gain of $2.3 billion.)
(Public Citizen also claims that Commerce Department data misleadingly inflates U.S. exports because it does not subtract re-exports, which are goods are shipped to the United States before being exported again, without alteration. Excluding re-exports, which the administration argues persuasively can distort trade figures, does not make much difference in the Korea numbers.)
Finally, one cannot view trade statistics in an abstraction. The U.S. economy has been the strongest in the world, while South Korea’s economy has slowed, thus hurting exports — not only from the United States but other countries as well, according to the Congressional Research Service. Imports from Japan, for instance, fell 12 percent from 2011 to 2013, almost double the drop in U.S. exports.
If you dig into the weeds of the ITC report, it says much of the gain in exports was expected in areas such as dairy products, meat products and apparel, most of which have shown real gains in the past three years. A more appropriate way to measure the impact of the Korea FTA would be to examine the impact of the trade in goods directly affected by the lifting of tariffs.
Finally, since the Korea FTA agreement was launched, manufacturing jobs in the United States have grown by 420,000, according to the Bureau of Labor Statistics. Is Public Citizen suggesting it would be more than 500,000 absent the agreement?
Lori Wallach, director of Public Citizen’s Global Trade Watch, and Ben Beachy, research director, emphatically insist that is not the case. They note that the group’s news releases contain careful caveats to show that they are simply applying the same ratio as used by the administration in calculating the initial 70,000 claim. “It’s a snapshot at three years,” Wallach said.
But that’s not really the case. The administration’s $150,000-per job was based on exports, and one cannot necessarily apply that to trade deficits. (Some imports are of products not produced in the United States.) “I would love for them to produce that ratio,” Beachy said, referring to a ratio on the impact of trade deficits on jobs, but since it is not available he said the export ratio is an appropriate substitute.
The Pinocchio Test
As we often say, two wrongs don’t make a right. While the White House’s math on 70,000 jobs was fishy, there is no reason for Public Citizen to try to replicate it — and then manipulate it even further.
As we noted, using the data on exported goods properly, without inflation adjustment or starting from a selective month, would yield a gain of 7,000 jobs. (We are not saying that is a correct number, just that that is what correct math would show.) While Public Citizen claims it is simply using the same ratio as the administration, they are applying a ratio for exports to trade deficits — which is not the same thing.
And while Public Citizen buries caveats in its news releases, that 85,000 number has been cited as an actual job-loss figure by lawmakers, making it even more irresponsible. We suggest the organization stop making this calculation — and rely on real facts if it wants to challenge trade deals. (Update: Economist Dean Baker says we gave too many Pinocchios, and Public Citizen posted an extended response. The Economic Policy Institute also offered a critique.)
Send us facts to check by filling out this form