Suppose Ford closes an assembly plant in Ohio and instead has its cars assembled in Mexico and shipped back to the United States. The workers in the Ohio factory have lost their jobs because of imports. This is a very simple point. For some reason supporters of trade deals like the Trans-Pacific Partnership (TPP) have trouble acknowledging this basic fact.
The difficulty that TPP proponents have acknowledging the jobs lost due to imports is bizarre, because the job loss does not mean that the TPP would be bad policy. It is simply a factor that must be assessed in considering the overall merits of the deal. It is not possible to have a serious assessment of the impact of TPP or any trade deal without considering the workers who would likely lose their jobs due to increased imports. It is also important to note that the impact stems well beyond the workers who lose their jobs to the much larger number who see a reduction in pay as a result of reduced demand for their labor.
With a recent report on the topic, the Congressional Research Service (CRS) seems to have joined the ranks of the denialists. The report goes to great lengths to argue that it is not possible to produce a figure for jobs lost due to imports that is comparable to the International Trade Administration (ITA) estimate that $1 billion of exports supports an average of 5,590 jobs. Rather than providing information to members of Congress about the likely impact of trade on jobs, this report seems to be a deliberate effort at obscuring the issue so as to leave members confused about the extent to which imports will displace jobs.
The report correctly notes that the ITA estimate was explicitly designed as an estimate of the jobs associated with $1 billion in exports, based on the goods and services the United States exports. It then argues at length that it is wrong to use this number for estimating the jobs lost due to $1 billion in imports.
While the composition of exports is different from the composition of imports, it is unlikely that the relationship between employment and output is hugely different between the export and the import competing sectors. In fact, since wages are somewhat lower in manufacturing in import competing sectors than in manufacturing in the export sector, it is likely that more job loss would be associated with a $1 billion increase in imports than would be associated with a $1 billion increase in exports.
It would be better to derive a number of jobs per billion dollars of imports directly based on the output of import competing industries (it is difficult to understand why the ITA has not done this), but in the absence of a number from ITA, the export number is a reasonable starting point. Economists recognize that it is better to use an inexact estimate than to have no estimate at all, which appears to be the position advocated by CRS.
It is worth noting that the average numbers calculated by CRS will be inappropriate for any specific policy or trade agreement. The vast majority of U.S. exports are to other wealthy countries like Canada, the European Union and Japan. Most of the increase in exports in the TPP will be to developing countries like Malaysia and Vietnam. For this reason, it is unlikely that the ITA number for average jobs per billion dollars for all exports will be very close to the relationship between jobs and any increase in exports as a result of the TPP.
In fact, for some exports there will be virtually no jobs impact. The increase in royalty payments and licensing fees as a result of stronger and longer patent and copyright protection may increase profits in the pharmaceutical and entertainment industries, but will have no direct impact on employment. And in the case of the closed Ohio assembly plant, all the factories that used to ship parts to Ohio are now producing exports, even though zero new jobs were created.
As the CRS reports notes, there will be problems calculating equivalent jobs for some imports since there may be no close domestic counterparts. But economists deal with these sorts of problems all the time. The most common method would be to look at industries that we think are closely related. For example, if it turns out that we don’t make boots in the United States, it would be reasonable to look at the relationship between jobs and output in shoes and other similar industries and apply this to the boot industry.
Rather than decrying the lack of a number from ITA relating imports to employment that Congress can use in assessing trade proposals, the CRS report seems to be celebrating it, as though it is better that members of Congress cast votes on trade in ignorance.
The CRS report is also misleading in implying that there is no link between total employment and the trade deficit:
“Some groups have equated bilateral trade deficits with a loss of employment. Most economists, however, argue that equating a trade deficit, whether on a bilateral basis or overall, with a specific amount of unemployment or job losses in the economy is questionable.”
While most economists would agree that there is no link between unemployment and the trade deficit when the economy is at full employment, most economists would agree that the economy has not been at full employment in the more than seven years since the housing bubble burst. For example, the most recent Budget and Economic Outlook from the Congressional Budget Office put the economy at about 2 percentage points below potential GDP, meaning that if there were more demand in the economy it could produce roughly $360 billion more output.
As long as the economy is below potential GDP, additional demand, whether from consumption, investment, government spending or net exports, will be associated with additional output and employment. As long as the economy remains below its potential level of output, most economists would argue that a larger trade deficit is associated with higher unemployment and a lower trade deficit is associated with lower unemployment.
There are many proponents of TPP who seem determined to avoid having an honest public discussion of the issues involved. CRS should not be in this camp.