Late last month, Trump started the Manufacturing Jobs Initiative, for which he will consult with major manufacturers to find ways to "best promote job growth and get Americans back to work again." Based on his own campaigning, Trump faults "bad trade deals" such as NAFTA, as a major culprit of why manufacturing jobs fled the country. For Trump, if we could just get better trade deals and close the trade deficit, we can "make America great again" by bringing back those manufacturing jobs. Throughout his campaign, Trump presented a nostalgic vision of manufacturing as it was in the 1950's, the one where a high schooler with a C-average GPA could go down to the factory, get a job, and have a middle-class lifestyle. I hate to burst the bubble here, but that world doesn't exist anymore. What's more is that I know that world isn't coming back, certainly enough to where I can call Trump out on a false campaign promise. How so?
Trump wants to bring manufacturing jobs back, but the problem is that he misdiagnoses. Trump wants to blame trade deals and China and Mexico on why there has been a decline of manufacturing employment. Gathering statistics from the Bureau of Labor Statistics, the St. Louis Federal Reserve shows the historic trends of manufacturing employment in the United States (see below). The number of jobs peaked in 1979 with 19.5 million manufacturing jobs, and has overall declined since then. In terms of manufacturing jobs as a percent of overall jobs, the United States reached that peak in 1943, and it has been declining ever since. The United States has been on an overall, decades-long decline when it comes to manufacturing employment, even with the increase since 2010 that was caused by advanced manufacturing. The news is better when we look at overall job creation, instead of just manufacturing jobs. Even with the overall decline in manufacturing jobs, data from the Bureau Labor Statistics show that the employment growth in the agricultural and service sectors from 2004 to 2014 more than compensate for the decline in manufacturing jobs. A similar decline has been taking place in manufacturing as a percent of GDP, and that is not just for the U.S., but also Germany nd the rest of the world (see World Bank data here).
With such a decline in employment and manufacturing as a percent of GDP, you would think that Trumps characterization of a dying manufacturing sector is accurate, right? Wrong! The chart below, which is from the Brookings Institution, says plenty. Contrary to Trump's narrative of a dying manufacturing sector, the manufacturing sector has gotten larger. As the Congressional Research Service illustrates in its January 2017 report on U.S. Manufacturing, the United States provide $2.17T value-added in manufacturing (CRS, p. 2). If manufacturing were on the decline, you wouldn't see that manufacturing accounts for 77 percent of research and development dollars spent in the U.S, and you wouldn't see over a trillion dollars in foreign direct investment (FDI).
Not only is the manufacturing sector larger, but it is more successful in terms of output. In 1980, it took 25 workers to generate $1 million in manufacturing output. Adjusting for inflation, it now only takes 6 workers to generate the same output (Brookings Institution). The United States has less manufacturing jobs, but produces more in manufacturing than ever. Since 1980, U.S. manufacturing exports have grown fourfold. This hardly sounds like a decaying manufacturing sector to me.
Why has output considerably outpaced employment? It is primarily due to productivity growth, and not trade policy. According to a 2015 study from Center of Business and Economic Research at Ball State University, 85 percent of job loss in manufacturing has been due to technological development, primarily automation. Even this more pessimistic study from MIT shows that trade policy only contributed to a quarter of the manufacturing employment decline, and that the remaining 75 percent was due to technological development. Automation has rendered millions of low-skilled manufacturing jobs redundant (see here for examples of automation). This is why Trump, or anybody else for that matter, can't bring the jobs back because, as the Wharton School of Business details, there's nowhere to bring them back from.
I covered automation in a blog entry a couple of years ago and discussed how robots are going to greatly shape the job market in the future. I concluded that a) while there will be job destruction, there will also be considerable job creation, and b) there are not going to be as many jobs lost as one would think. Also, Trump plays up the lost jobs while ignoring the creation of new jobs or value that consumers derive as a result of the automation. In 1950, the United States went from having 20 million to just 3 million people living on farms, but I don't hear any complaints about higher salaries, lower food prices in inflation-adjusted dollars and percent of overall individual budget, or working less hours as a result. A similar displacement happened with the rise of computers. Yes, there was a loss in administrative jobs and jobs in other sectors, but who is willing to trade in their laptops and smartphones to go back to those times?
Unless Trump decides to ban automation or other technological progress, nothing that Trump or anyone else can do is going to reverse the trend towards automation. To illustrate just how gone these jobs are, think of this: Since 1980, the United States has lost 6.4 million, or about one third, of its manufacturing employee base. To recover just half of those lost jobs, the Brookings Institution calculates that Trump would need to increase productivity by an unprecedented 26 percent. How we approach job creation needs to be a forward-looking endeavor instead of lamenting over what used to have.
What this means is that Trump can't be blaming China or slapping tariffs on Mexico, especially given both the negative effects of tariffs and positive effects of freeing up economies. If Trump wants to properly respond to the increased automation in the manufacturing sector, he cannot halt progress, but instead will need to find ways to adapt to the not-so-new reality of manufacturing. As the Economist points out, advanced manufacturing provides good jobs, but these jobs need skills and adaptability. With the increasing prevalence of automation, if Trump wants people to work in manufacturing, it is going to take a ton of industry-related training. The Brookings Institution illustrates the importance of revamping education to train students in science, technology, engineering, and mathematics (STEM). However, given the decline in manufacturing employment, both in absolute numbers and percent of overall employment, Trump will need to focus on more than just manufacturing. He should focus on making sure people have the education to take on high-skilled jobs in the service sector, as well. He should focus on some of his other more positive campaign promises, whether it is to lower the corporate tax or to remove burdensome regulations on businesses. If Trump focused less on protectionist policies and instead focused more on liberalizing the economy to encourage further investment in the United States, there is a more-than-distinct chance that he can indeed make the American economy great again.