Michael Hicks is the George and Frances Ball Distinguished Professor of Economics and the director of the Center for Business and Economic Research at Ball State University. His column appears in Indiana newspapers.
Manufacturing employment has enjoyed a long recovery since the darkest days of the Great Recession. As of late last year, we have a full 108,000 more factory jobs than in summer 2009, which marked the trough of the business cycle. This recovery eased some of the deep impacts of automation and trade that cost the United States and Indiana about one third of all factory jobs. Here in Indiana, from January 2000 through the start of the Great Recession, factory employment dropped some 126,600 workers. From the December 2007 through the end of the Great Recession in July 2009, factories shed a further 119,700 jobs. This employment loss was a full 36.8 percent of all factory jobs in Indiana.
There is an interesting debate among economists about just what caused those factory job losses. The consensus appears that the majority of job losses in factories were due to productivity gains. However, much of the observed increase in productivity likely came from businesses responding to significant threat from foreign competition. It’s not clear how those job losses should be accounted for, but there are a few facts that bear on the discussion.
First, growth in transportation and logistics jobs has more than offset the losses in manufacturing, and so has growth in other sectors. International trade doesn’t cause a net loss of jobs, but changes the skills and location of jobs. Second, trade deficits and deals are not correlated with large factory job losses. The last two lengthy periods of factory job growth occurred in the years after NAFTA and following the Great Recession. These were two periods of growing trade deficits. However, the big factory job losses of the early 2000s occurred at a time of both rapid growth in our trade deficit and very rapid growth of factory productivity.
From 1990 through 2007, productivity grew by 4.0 percent each year. At that rate of growth, by 2017 51 workers could make what took 100 workers to produce in 1990. Even allowing for measurement error, this alone can account for most of the factory job losses. And in that fact lies a warning for today. Since the beginning of the recession, factory productivity has grown at less than 1.0 percent per year. More worrisome, since 2010, factory productivity has averaged -0.1 percent. So, the average factory worker produces less in inflation-adjusted terms today than they did in 2011.
The declining productivity since the end of the Great Recession has caused firms to hire more workers to meet the demand for their products. Now, it may be that US factories grew so quickly in the 2000s that the current productivity slump is related simply to a period of ‘catching up.’ Such periods of fast and slow productivity growth are common. But, it also means that if productivity growth again picks up, US factories will return to their four-decade-old trend of shedding jobs.
One way to gauge business expectations about productivity is to examine the composition of jobs. If there has been a shift in the types of factory workers, this might hint at the type of productivity gains factories anticipate in the future. If there is a big shift in the education composition of employees, firms are likely expecting future productivity growth. If factories are hiring the same types of workers, we should expect much fewer future productivity changes.
The most recent data we have suggests we are in the midst of a significant shift in the labor force composition within manufacturing firms. From 1998 to the end of the Great Recession, we lost 11,600 factory workers with a college degree, 24,300 with some college-level training and a whopping 62,000 who have not been to college. This last category includes non-high school graduates and those with post-secondary certifications not received in a college setting.
Since the end of the recession, through early 2018, we’ve gained back many of these jobs. We’ve made up 68 percent of the college graduate jobs, 56 percent of the jobs requiring some college, and 59 percent of the non-college jobs. But, this last category is deceiving. As it turns out, we’ve actually created more jobs for workers without a high school diploma than we have for those who graduated. This includes workers with postsecondary training outside a traditional college. This is worth restating. During this lengthy period of factory job growth, the one set of workers who have not enjoyed a resurgence of hiring are those with a high school degree, including those with certificates, but have not been to college.
It is not hard to draw some conclusions from this. Since 1998 factory production in Indiana has risen 13 percent, while employment declined by more than 18 percent. Within Indiana factories, productivity growth stalled after the Great Recession, so firms had to hire more workers to meet growing demand for goods. These new workers are disproportionately high-skilled and low-skilled workers. The mid-skilled workers, including those who avail themselves of non-college training, are a rapidly declining share of the factory labor force.
The policy lessons should be obvious. Future productivity growth will likely prove a panacea to the better skilled workers. However, that very same productivity growth will make redundant low- and middle-skilled factory workers. This is the message of three decades of factory employment changes. In a better world, these facts and conclusions would inform the flow of dollars we invest in children and workers. We’d be wise to spend more on developing those skills that ready more children for advanced post-secondary education.