October 11, 2016
Given his history of promoting racism, xenophobia, sexism and his recently exposed boasts about sexual assaults, not many people want to be associated with Donald Trump. However, that doesn’t mean everything that comes out of his mouth is wrong.
In the debate on Sunday Donald Trump made a comment to the effect that because of Nafta and other trade deals, “we lost our jobs.” The NYT was quick to say this was wrong.
“We didn’t.
“Employment in the United States has increased steadily over the last seven years, one of the longest periods of economic growth in American history. There are about 10 million more working Americans today than when President Obama took office.
“David Autor, an economist at M.I.T., estimated in a famous paper that increased trade with China did eliminate roughly one million factory jobs in the United States between 2000 and 2007. However, an important implication of his findings is that such job losses largely ended almost a decade ago.
“And there’s no evidence the North American Free Trade Agreement caused similar job losses.
“The Congressional Research Service concluded in 2015 that the ‘net overall effect of Nafta on the U.S. economy appears to have been relatively modest.’”
There are a few things to sort out here. First, the basic point in the first paragraph is absolutely true, although it’s not clear that it’s relevant to the trade debate. The United States economy typically grows and adds jobs, around 1.6 million a year for the last quarter century. So any claim that trade has kept the U.S. from creating jobs is absurd on its face. The actual issue is the rate of job creation and the quality of the jobs.
Here there are three issues to consider.
1) The direct job loss — the jobs that were displaced due to imports substituting for domestically produced goods and services;
2) The wage effects — the downward pressure on the wages of workers that retain their jobs that can result from job loss and also the threat of job loss;
3) The impact of a trade deficit on the level of demand in the economy.
Taking these in turn we now have some pretty solid evidence on some of the job loss attributable to trade. David Autor’s work found that imports from China cost the economy more than two million jobs in the years from 2000–2007.
“Estimates of the net impact of aggregate demand and reallocation effects imply that import growth from China between 1999 and 2011 led to an employment reduction of 2.4 million workers” (page 29).
These are workers who are directly displaced by import competition. In addition, as the article goes on to note, there were more workers who likely lost their jobs to the multiplier effect in the local economies most directly affected by imports.
The impact of trade with China was more dramatic than trade with Mexico and other countries because of the huge growth in imports over a short period of time. However, even if the impact from trade with other countries was smaller, it still would have a substantial effect on the communities affected.
It is also worth noting that even though our trade deficit has declined from its 2006 peak (the non-oil deficit has recently been rising again), workers are constantly being displaced by imports. The Bureau of Labor Statistics reports there have been an average of 110,000 layoffs or discharges a month in manufacturing thus far this year. If just a quarter of these are trade-related, it would imply that more than 300,000 workers a year are losing their jobs due to trade.
Of course people lose jobs for other reasons also, like increased productivity. So the fact there is job loss associated with trade doesn’t make it bad, but it is not wrong to see this as a serious problem.
The second point is the wage effect, which can go beyond the direct impact of job loss. The oil market can give us a useful way of thinking about this issue. Suppose that Saudi Arabia or some other major producer ramps up its oil production by one million barrels of oil a day. This will put downward pressure on world prices, which will have the effect of lowering prices in the United States as well. This could mean, for example, that instead of getting $50 for a barrel of oil, producers in North Dakota will only get $40 a barrel. This will mean less money for workers and companies in the oil industry. In the case of workers, it will mean fewer jobs and lower pay.
This can happen even if there is very little direct impact of trade. The increased supply of Saudi oil may result in some modest reduction in U.S. exports of oil, but the impact on price will be much larger. The analogous story with trade in manufactured goods is that the potential to import low cost goods from Mexico, China, or other countries can have the effect of lowering wages in the United States, even if the goods are not actually imported.
Kate Bronfenbrenner, a professor of industrial relations at Cornell, documented one way in which the potential to import can have the effect of lowering wages. She found that employers regularly used the threat of moving operations to Mexico as a way to thwart unionization drives. While most workers are not typically involved in unionization drives, it is easy to imagine this dynamic playing out in other contexts where employers use the real or imagined threat from import competition as a reason for holding down wages. The implication is the impact of trade on wages is likely to be even larger than the direct effect of the goods actually brought into the country.
Finally, the balance of trade will have an impact on the overall level of employment in the economy when the economy is below its full employment level of output. Until the Great Recession, most economists did not think that trade could affect the overall level of employment, but only the composition. This meant that trade could cause us to lose manufacturing jobs in the Midwest, but these job losses would be offset by gains in Silicon Valley and other tech centers. This could still mean bad news for the manufacturing workers who lost their jobs, but the net effect for the country as a whole would still be positive.
The Great Recession changed this view, as many economists came to believe that the United States is facing a period of secular stagnation: a sustained period in which lack of demand in the economy constrains growth and employment. In this context, the trade deficit is a major cause of the lack of demand since it is spending that is creating demand in other countries rather than the United States. If we could reduce the annual trade deficit by $100 billion then as a first approximation it will have the same impact on the economy as a stimulus of $100 billion.
From this perspective, the trade deficit is a major source of job loss. Our current trade deficit of $500 billion a year (@ 2.8 percent of GDP) is a major drag on demand and employment. For this reason, a politician would be absolutely right to cite trade as a big factor in the weakness of the labor market.
It is worth noting that many economists (including many at the Federal Reserve Board) now believe that the economy is close to its full employment level of output, in which case trade is not now a net cause of job loss even if it had been earlier in the recovery. There are two points to be made on this view.
First, there are many prominent economists, such as Paul Krugman and Larry Summers, who argue that the economy is still well below its full employment level of output. So, this is at least a debatable position.
Second, if we accept that the economy is near full employment it implies that close to two million prime age workers (ages 25–54) have permanently left the labor market compared to 2007 levels of labor force participation. (The gap is close to four million if we use 2000 as our comparison year.)
There is no generally accepted explanation as to why so many prime age workers would suddenly decide they didn’t feel like working, but one often invoked candidate is the loss of manufacturing jobs. The argument in this story is that the manufacturing sector provided relatively good paying jobs for people without college degrees. With so many of these jobs now gone, these workers can’t find jobs. If this argument is true, then it means that trade has cost the country a large number of jobs even if the economy is back at full employment.
In short, there are good reasons for a politician to complain about trade as a major source of our economic problems. There is much research and economic theory that supports this position.
Comments