Through rose-colored glasses.

Photographer: Luke Sharett/Bloomberg

Where Have All Our Wages Gone?

Megan McArdle is a Bloomberg View columnist. She wrote for the Daily Beast, Newsweek, the Atlantic and the Economist and founded the blog Asymmetrical Information. She is the author of "“The Up Side of Down: Why Failing Well Is the Key to Success.”
Read More.
a | A

Many theories have been advanced for why unions, and median wages, aren't growing very fast. Some say there's a causal link, which runs something like this: The Reagan administration gutted union protections, making it harder to organize workers. Without a powerful union to represent them, workers were at the mercy of greedy bosses who ruthlessly forced down their wage packets. What America needs, therefore, is a powerful labor movement, protected by more powerful laws that favor organizing of employees.

Previously: Unions Are Powerless. Workers Aren't.

It's an obviously attractive story, both because it gives us a nice, satisfying morality tale and because it offers a (relatively) easy policy solution. But there's a little problem with this story: How, then, do you explain the fate of the United Automobile Workers employees at General Motors? Theirs was a very powerful union, one that managed to stave off a lot of job-threatening changes. And thanks to that power, the union was able to mobilize politicians to get them a much better deal out of GM's bankruptcy than they probably would have gotten in a more normal proceeding. What they weren't able to do was save the old wage structure. Old workers still have some semblance of the old package, but new workers make substantially less. Meanwhile, there are a lot fewer of them than there used to be. In the 1970s, GM employed nearly a half million auto workers. Today, that number is more like a tenth of that.

What happened? The same thing that happened to workers across this great land of ours: competition from machines and competition from abroad. GM now has robots doing a lot of the work that used to be done by hand. It also faces a lot more competition than it did in those happy days when the Big Three automakers had virtually total control over the U.S. market. The company that once sold half of all the cars on American roads now commands less than 20 percent of the domestic market.

Organizing a quasi-cartel like the mid-century automakers is a very different project from organizing a company that competes with companies from all over the world. When you're dealing with a cartel, it's easy to make big wage demands, because the companies can just pass those demands on to consumers, either by raising the price or taking some quality out of the product. If the consumers don't like it, well, they can walk. But when you open up that market to competition, consumers have another choice: They can buy a product that isn't produced by your union. Your company loses market share, your members lose jobs. 

In a competitive market, the price is the price is the price. You can rail against it, you can jawbone people about it, you can complain that it should be otherwise. What you cannot do is change it. If you try to force the price above the market-clearing level, you can do so only by sacrificing sales. GM managed to hide that fact from itself for a while, because it lowered the quality of its cars rather than raise the sticker number. That let it run things out a while longer, though at the price of eroding its brand. When the market figured out that the cars weren't as good, the price became the price again, and its market share fell accordingly.

Ironically, the International Longshore and Warehouse Union is a beneficiary of the same forces that are killing unions everywhere else: Its workers are the folks who unload the stuff that comes in from abroad. Those workers are sitting right at the bottleneck through which all that foreign competition flows to the U.S., and due to some smart planning by the union in earlier decades -- such as insisting that the West Coast ports would bargain as a bloc so that they can't play different locals against each other -- they can cork that bottle anytime they want. That gives them the ability to extract a little tariff on all that trade. Their rake-off is a tiny fraction of the overall flows, but that still translates into substantial sums for the few thousand workers who are holding onto that cork. The more trade, the better the concessions the ILWU is able to extract -- and the worse workers in other unions do.

There's a notable exception to this, of course: government workers. Government unions do not have to worry about having their jobs outsourced to China. But they do have to worry about political support for their activities, as Governor Scott Walker's adventures in Wisconsin show.  

Where I sit, in Washington, these proceedings were viewed with astonishment. Scott Walker was beating up on teachers -- underpaid, overworked community stalwarts! In a progressive state such as Wisconsin, surely the populace would rise up and show the Republican Party that while its tactics might work in the backward provinces of the Deep South, the sensible north would have none of such nonsense. Instead, he beat the protests, the legislative stalling tactics and, finally, a recall election.

What happened? Here's my working theory: The disparity between government and private-sector outcomes had become too great. People in a coastal city have trouble imagining this, but in smaller places -- particularly rural areas --teachers are relatively affluent. Their salaries are good by community standards, they can't be fired, their benefits are outstanding, and they get three months off a year -- which, no, they do not all spend on 12-hour-a-day lesson planning, and if you're going to insist that they do, then no one in your small town had better see a teacher at the shopping center or the community pool in the middle of the day. People do like their teachers, and they will get mad if you speak ill of them. That doesn't mean that they're willing to see their taxes increased to cover pension deficits and guaranteed salaries and gold-plated benefits, not when they themselves had to cancel the vacation and pull Junior out of travel hockey.

It's all very well to say that instead of punishing teachers, we should give everyone else the same job security, benefits and so forth -- and, well, lots of people said that. The problem with that is that no one proposed a realistic plan to do so in the face of foreign competition and automation. The price was winning. And that, in turn, lowered the price people were willing to pay for government workers in Wisconsin. Or for health care. Or for other sectors, shielded from competition, where unions have managed to hold on.

If you're not depressed yet about the future of wages, here's the real kicker: Trade doesn't just sort of evenly depress wages across the board. It has favorite sectors that it likes to pick on. One of those sectors is agriculture, and OK, it's not like we all pine for the days of those fantastic migrant worker jobs that used to let a man support a middle-class family in style. But the other sector it really likes to pick on is manufacturing. And manufacturing jobs tend to have higher productivity than service work, particularly for less skilled workers. 

According to my handy Bloomberg terminal, GM's U.S. operations, with around 50,000 hourly employees, generate $85 billion in revenue. Wal-Mart's U.S. operations, with around a million hourly employees, generate about $330 billion. Much more revenue -- but to make four times as much money, Wal-Mart needs 20 times as many front-line employees. Of course, this is necessarily a very crude metric, because more of Wal-Mart's workers are part time, and it also takes more in the way of inputs and capital to make a car. Yet no matter how you refine those numbers, you cannot make them add up to auto-worker wages for retail work.

You often hear that U.S. manufacturing is in decline, which is incorrect. U.S. manufacturing output is doing splendidly. What's suffering is U.S. manufacturing jobs. And the jobs available in other sectors for lower-skilled workers don't pay as well as the old high-productivity manufacturing jobs. You just can't generate massive economies of scale in fast food or hairdressing.

We should be careful about getting too nostalgic for the old manufacturing jobs, which did indeed pay well but were also pretty awful. I've spent some time looking at assembly lines for work, and they always inspire two thoughts: "Wow, how amazingly productive all this is!" and "Wow, I would kill myself if I had to spend the rest of my life doing this." Line work is monotonous in a way that even the proverbial crappy retail job is not. In fact, I am struggling to think of any other sort of work that is so particularly ill-suited to the human psyche. This was something that everyone knew about these jobs right up to the point where they started going away and we got all misty-eyed for a life spent riveting the same four bolts into place on a car frame.

But all that said, the wage story is incredibly important. And it will become more important as we lose other sorts of jobs to trade or automation, from the already-disappearing secretaries to the paralegals and call-center operators who get outsourced to Bangalore. The work may not have been thrilling, but it provided a decent paycheck and some stability for people, some of whom are now being forced into the sorts of low-wage, high-turnover work that used to be done by folks who weren't trying to survive on the paycheck, such as teenagers and housewives.

If you think that unionization produces big bumps in wages, we need to add another wrinkle to this story: Manufacturing is easier to unionize than many other sorts of operations. These workers are concentrated in large production plants, doing routine tasks that are relatively easy to describe and to arbitrate when disputes arise. (Did, or did not, that worker rivet his four bolts 150 times an hour?) It's relatively easy to set up a promotion system by some objective criteria, such as seniority, that can be written into a collective bargaining contract; intangible factors, such as how good this person is at customer service, don't really enter into the decision. Service work tends to be more distributed, meaning you have to organize lots of sites, and less routine, meaning it's harder to agree on job descriptions and promotion arrangements, and harder to agree whether criteria have been met. Oh, and because it's less productive, there's less value for the unions to claim for workers -- meaning less surplus available to pay union dues.

That's why we can see rising wages at Wal-Mart and still see median incomes looking kind of floppy. Wages could rise across every sector, but if the available jobs are shifting out of high-productivity sectors and into lower-productivity industries such as retail, wage growth will continue to be anemic.

Wednesday: The future of manufacturing.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

To contact the author on this story:
Megan McArdle at mmcardle3@bloomberg.net

To contact the editors on this story:
Brooke Sample at bsample1@bloomberg.net
James Gibney at jgibney5@bloomberg.net