From the FT:
The US Chamber of Commerce has claimed that the South Korea deal alone will create 280,000 jobs, while the White House has more modestly suggested a net 70,000 gain.
However, the Economic Policy Institute, a left-leaning think-tank partly funded by labour unions, has said that 159,000 manufacturing jobs will be lost.
So there's a business group claiming that the U.S.-Korea FTA will create 280,000 jobs, and a liberal think tank arguing that it will cause a loss of 159,000 jobs. The number in the middle of these two figures is a 60,500 job gain, which is not far from the White House estimate of a 70,000 gain. Did the White House just split the difference between the two, and add a few jobs for good measure?
All of this suggests to me that calculating the employment effects of trade agreements is not particularly scientific.