Manufacturing in the United States Since the Year 2000 – Where Did the Jobs Go?

The first in a series about the future of manufacturing in the United States.

The numbers are sobering. Manufacturing employment in the United States has dropped from 17.2 million workers in the year 2000 to 12.3 million in 2015. That’s despite a gain of almost a million manufacturing jobs since the 2008-2009 recession.

It’s a big deal. In a time when income and wealth inequality are a growing issue, a shrinking American manufacturing sector means millions will no longer be able to pursue jobs that once offered one of the best tickets to middle-income security and the American Dream.

There’s lots of conventional wisdom around this topic. One of the oldest arguments? Unions screwed-up American manufacturing. Excessive government regulation is also blamed – the job-killing impact of employment, environmental and safety rules that other nations don’t impose or enforce. Another, more twenty-first century theory is that all those jobs went overseas – the outcome of trade agreements that have favored countries like China and Mexico.

The truth is a lot more nuanced, and this series of posts will look at the both the conventional wisdom and some of the real, underlying root causes.


The Usual Suspects – American Labor Unions

In this post – the role unions have played in the manufacturing sector. The below chart tracks total US manufacturing employment and sector union representation on the top, and on the smaller, lower chart work stoppages (in general, strikes) involving more than 1,000 employees.

Big strikes are a reasonable, albeit ill-defined measure of unions’ negative impact on management decision making about jobs. The more strikes, the more management will be disinclined to expand employment in unionized facilities. (And a bit more of a disclaimer – this Bureau of Labor Statistics data includes all work stoppages from all sectors of the economy, so stoppages in manufacturing are only a portion of the data shown here.)

Union Membership

Union membership – both in terms of absolute numbers and as a portion of the total manufacturing work force – has been on a downhill toboggan run since the 1980s. Accompanying that has been a continuing decline in labor unrest to the point that major strikes have become a rare, almost extinct species in 21st century America.

(There’s no doubt that some of this labor “peace” reflects hard, post-2000 economic realities. In a world of collapsing manufacturing employment and volatile economic times, a strike by manufacturing workers looks a lot like suicide.)

Another point – the portion of the manufacturing sector using non-union employees grew steadily in the 1980s and 1990s. It offset the loss of union jobs so that, in rough terms, overall manufacturing sector employment was flat during the 1980s and 1990s.


Higher Union Wages – A Smoking Gun?

What about higher union wages? Could they have driven the job losses in the 2000 to 2015 period?

Unionized manufacturing employees make a lot more than their counterparts when total wages and benefits are calculated. Bureau of Labor Statistics estimates for 1987-2001 show blue collar, unionized employees in the manufacturing sector made anywhere from 35% to 50% more than their non-union counterparts. While wages were somewhat higher, benefits (such as health insurance and pensions) were the primary source of higher overall compensation. This delta has not dramatically changed over the past 15 years.

Did these wages have something to do with the fall-off in unionized employment from 1980 to 2000? Absolutely. Higher union wages, lousy products, 1950s-vintage production engineering and savvy overseas competition contributed to permanent job losses in the US automotive industry. (The UAW’s membership alone has dropped by 1 million workers since 1979.) The US steel industry has suffered a similar fate. There are other, less dramatic examples.

Since 2000, job losses in manufacturing have hit both non-union and unionized workers alike. Non-union job losses in manufacturing totaled 3.5 million employees since 2000, or about 25% of the total number of non-union employees. On the union side, there were 1.3 million job losses totaling over 50% of the unionized manufacturing employees.

The numbers suggest a role unions played in these manufacturing sector job losses. A crude estimate is to start with the debatable assumption that the only reason union losses were proportionally higher than non-union losses was higher union wages and benefits. That amounts to 650,000 jobs, or about 12% of the total number of union and non-union 4.8 million manufacturing jobs lost between 2000 and 2015.

Any labor economist that can fog a mirror would laugh at that calculation – the real world is a lot messier. But it puts a range on the possible impact of higher union compensation – it probably somewhere around 10%.


American Manufacturing Unions in the 21st Century – a Death Spiral?

In short, higher union wages and benefits probably played a minor role in 2000 to 2015 manufacturing sector job losses.

This makes sense when taking the long view about American manufacturing unions. Since 2000, they’ve represented fewer and fewer workers. As their numbers have declined (accelerated by two recessions), emboldened management at both large and small manufacturers have negotiated tougher contracts and been less constrained in making employment decisions. With a few notable exceptions (e.g. the major US automakers), there isn’t a manufacturing company in the country that can’t set up a non-union factory somewhere in the United States. (An illustrative example of how companies can expand non-union employment is Boeing’s new aircraft assembly plant in South Carolina.)

Looking ahead, there’s little reason to believe US manufacturing unions will see their membership grow or their economic influence expand. They may well be a death spiral in which declining membership weakens their long term negotiating position to the point where there will be little economic incentive to join or support a manufacturing union in the US.


If Not Unions . . .

So if unions aren’t the culprit for the manufacturing job losses over the past fifteen years, who is? Follow-up posts will explore this and how policy changes could create a renaissance in US manufacturing.

Agree? Disagree? Join the discussion.