Last week, as Donald Trump was trying to distract attention from his impeachment trial, he was holding events touting his big trade victories. The two items for celebration were the new NAFTA, dubbed by Trump as the U.S.-Mexico-Canada Agreement, and a “phase one” trade deal with China. While these deals may be useful props for the impeachment distraction, they are unlikely to offer much to the manufacturing workers who Trump claims are at the center of his trade agenda.
The new NAFTA may lead to some modest shifts of employment in the auto industry, but its impact on the larger manufacturing sector will be invisible for all practical purposes. The China pact includes bizarre commitments of Chinese purchases of specific items. It’s not clear how these would be enforced, but if they choose to comply with the treaty and play games, there is nothing that prevents China from meeting its commitments by being an intermediary between the United States and other importers of U.S. goods.
Specifically, if China needs to buy another $50 billion of U.S. manufactured goods to meet its commitments, it could simply make $50 billion in purchases and then sell them to current U.S. importers in Japan, Korea, and elsewhere. That leads to no net gain in U.S. exports, but would comply with China’s trade commitments.
The markets have already indicated that they are not impressed with the impact of this deal on U.S. exports. The price of major farm commodities, like soy beans and wheat, remain well below their levels earlier in the decade. If there was an expectation that China’s purchases were going to substantially improve the fate of America’s farmers, then we should have seen the price of these products soaring after the outlines of the deal became known.
Trump punted on the one issue that likely would have made a substantial difference in the U.S. trade balance, the value of the dollar. After running around the country for two years denouncing China as a “world class currency manipulator,” the deal does nothing to change the exchange rate between the Chinese yuan and the dollar. (The term “manipulation” implies something which is surely not appropriate in this case. China openly manages the value of its currency, so we don’t have to catch them doing something secretly in the dark.)
Anyhow, it will take some time to see how these deals fully play out in practice, but we do have plenty of data to assess how manufacturing workers have been doing thus far under Trump. The answer for the most part is not very good.
At the most basic level, employment in the sector has been virtually flat since July, rising by a total of just 5,000 over this period. Manufacturing employment fell 12,000 in December. The situation looks considerably worse if we look at hours worked. The index of aggregate hours worked in the sector peaked in December of last year and has dropped 0.5 percent over the last year.
The wage picture looks even worse. Wage growth in manufacturing has consistently lagged overall wage growth in the private sector under Trump. The average hourly wage for manufacturing workers has risen by less than 2.3 percent annually over the last three years compared to almost 3.0 percent for the private sector as a whole. In fact, the average hourly wage for the private sector as a whole exceeded the average for manufacturing for the first time ever in May of 2018.
The situation looks even worse if we look at the weekly wage. While average weekly earnings have risen an average of 2.5 percent in the Trump years, in the last year they have risen just 1.3 percent. This means that the real weekly wage for manufacturing workers had fallen by roughly 1.0 percent over the last year.
While manufacturing has been hit hard by competition from low paid workers in the developing world, it has also been hit by a plunge in the unionization rate. The biggest hit was from 2000 to 2007, as the trade deficit exploded. The country lost nearly 40 percent of unionized workers in manufacturing in these years.
The number of union members in manufacturing fell further in the Great Recession. It then recovered somewhat as employment expanded, but peaked in 2013 and has been headed in a mostly downward path the last six years. The number of union members in manufacturing in 2019 was just over 46 percent of the number in 2000. The 8.7 percent current unionization rate in the sector is somewhat above the private sector average of 6.4 percent, but below the economy-wide average of 10.3 percent.
With this sharp drop in unionization rates, it is not surprising that manufacturing workers have not been able to secure wage gains in step with the rest of the workforce. The few new jobs that have been added in manufacturing in this recovery have mostly not been union jobs or especially well-paying.
This has meant that the industrial Midwest, which is still more heavily unionized, has largely been bypassed by the growth in manufacturing employment during the Trump years. Here is the story for the five Midwest states that were considered swing states in the last election.
Change in Manufacturing Employment
Last Year Since January 2017
Michigan -6,100 +9,800 (1.5 percent)
Minnesota -1,900 +1,900 (0.6 percent)
Ohio -2,200 +12,100 (1.8 percent)
Pennsylvania -4,500 +3,700 (0.6 percent)
Wisconsin -6,000 +4,100 (0.9 percent)
Source: Bureau of Labor Statistics.
By comparison, manufacturing employment for the country as whole is up 487,000 since Trump came into office, an increase of 4.0 percent. This means all of these states have seen gains in manufacturing far below the national average.
The prospects for the immediate future do not look much better. The Federal Reserve reported that manufacturing production increased a modest 0.2 percent in December, but that still left it 0.3 percent below its August level and 1.3 percent below its year ago level. Assuming even modest rates of productivity growth, production would have to increase by 1.0-1.5 percent annually just keep employment levels constant.
Other labor market measures also indicate weakness in manufacturing, most notably data on job openings. According to the Bureau of Labor Statistics, the rate of job openings in manufacturing was 2.9 percent in November, down from 3.8 percent last year. The one-month diffusion index, which reports the percentage of industries in the sector that expect to increase hiring next month, was just 44.7 last month. That compares to a reading of 65.1 in December of 2018 and 71.7 in December of 2017. These measures certainly suggest that a hiring boom in manufacturing is not imminent.
In fairness, it is difficult to know at this point how Trump’s trade deals, especially the one with China, will play out. However, to date, Trump certainly has not been able to produce big gains for the manufacturing industries who he had promised to help in his campaign. The trade deficit has actually risen as a share of GDP in the first three years of his administration.
The major impact of his trade war to date has been to reduce manufacturing output and investment by both raising costs with his tariffs and increasing uncertainty about future trade arrangements. Investment in equipment and structures has been falling the last three quarters. There is still some time where things could turn around before the election, but with each passing month this is looking less likely.
This article first appeared on Dean Baker’s Patreon page.