That Pesky Skill Shortage in Manufacturing
A closer examination of the current shortage of machinists in the United States yields some interesting findings, especially when compared to a similar reported shortfall in 1982.
You've probably noticed we are still hearing these skill-shortage stories. There is a common theme to these: The reporter almost always visits a small manufacturer who says they can't find anyone to hire. Then they go on to say the problem is that there is no one out there who can do the job, and the reason for that must be that the schools are failing. This was the story line of 60 Minutes (where I made a brief appearance) and another recent column by Tom Friedman in the New York Times.
Because these stories are inevitably about small manufacturing operations and virtually always about jobs associated with machinist skills, I thought it might be interesting to look into what is happening in that job market to understand these complaints about not being able to hire. The employers, no doubt, have some insights into the problem, but just as asking a job seeker why he or she can't find a job may not always lead to the most objective answers, asking employers why they cannot hire workers might not always lead to the most objective responses either. Neither group is inclined to look at what they're doing as a possible source of the problem.
Here's what I found.
First, the number of machinist jobs in the United States has declined quite a bit over time, more than 20 percent in the past 20 years, while the overall number of jobs in the economy has grown by 40 percent. Going forward, the Bureau of Labor Statistics projects that growth in these jobs will be below the average for the economy. This has been a declining field.
Second, what has happened to wages? An obvious explanation as to why it might be difficult to hire is that wages are not high enough. Just as a reminder from Economics 101, if you can buy what you want were you to pay more for it, then that means there is no shortage. The BLS calculates that the average pay for a machinist in the United States now is just above $40,000. Not bad, depending on your standards. What was it in the past? In 1982 -- also a recession year -- the average annual pay for a machinist in 2012 dollars was $48,000. So in real terms, the average pay for a machinist has fallen by 20 percent.
Third, what's happened to skill requirements? Machinist jobs used to be seen as the highest-skilled occupations in manufacturing. At the very least, machinists needed to be able to use basic trigonometry and algebra, sometimes very basic calculus. High-school math, but still math. We hear about how manufacturing jobs are different now, how they require a lot more skill and especially more math than they did before. Is that true for machinists?
I spent some time searching websites where machinists talk about their jobs, especially how they have changed over time, and here is what they say: The big change is the rise of computer-assisted machining tools. Among other things, these tools will do all the math calculations that machinists in the past had to do on their own. Some of the machinists say that it is still useful to know how to do those calculations, while others say there is no real need to know. Yet none I read say that it is better to do them yourself than to let the machine do it. (If this sounds wrong to you, ask yourself how often you use paper and pencil to add up numbers rather than use a calculator, or even to check the calculator's answers.) It is hard not to believe that the amount of math that the individual machinist needs to perform has declined.
Are there new requirements? Yes. Those that involve interacting with the computer-based machines. Depending on the task and the choices made by employers as to how to divide them between engineers and machinists, the new tasks may require programming the machines, usually simple programming by IT standards, or just operating the software, the way white-collar workers might use an Excel spreadsheet.
Let's say that the real skill issue comes because machinist jobs require some real programming skills.
One obvious problem is that programming jobs pay almost twice as much per hour. If employers really need machinists with programming skills, it's going to be hard to hire them paying that much less.
The simplest explanation for any perceived difficulty in hiring machinists -- or hiring people to become machinists -- is that it is just hard to attract people to a field that is in relative decline, where wages are much less than they have been in the past, and where the skills needed are useful in better-paying jobs.
At the same time, we know that traditional employer/union-apprenticeship programs that provided a path into machinist work have largely collapsed. That's got to be part of the problem, as well.
Here is something else I learned, from looking at 1982. A BLS study investigating employer complaints about a shortage of machinists that year found no real evidence of a skill shortage. The author, Neal Rosenthal, suggested that the way employers will address any difficulty in hiring would be to do what they have always done: raise their wages, train more new hires (he noted that formal apprenticeships had declined) and invest in productivity-enhancing equipment that reduced the need for machinists. Sounds like good advice now. All three approaches have the advantage of being good for the economy and society.
Finally, it has struck me how many of the complaints about not having anyone to hire come from businesses that essentially have no HR function. There's no one there with the skills to hire, to develop talent internally, or even to assess the market to know what to pay.
Maybe that's the real skill shortage.
Peter Cappelli is the George W. Taylor Professor of Management and director of the Center for Human Resources at The Wharton School. His latest book is Why Good People Can't Get Jobs: The Skills Gap and What Companies Can Do About It.