You can’t blame legislation for inequality

By Felix Salmon
June 12, 2012
NYT today, I review the new books from Paul Krugman and Tim Noah.

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

In the NYT today, I review the new books from Paul Krugman and Tim Noah. Short version: these are both really smart people, whom you should pay attention to. But neither has produced a great read, a book which works really well qua book. Krugman and Noah have good reason to be upset at me for this, because the fact is that the overwhelming majority of the nonfiction books which pile up around me every day are just as dull as theirs are. To say that these books are a bit of a slog is not to say something mean about Krugman and Noah in particular, so much as it is to say a simple truth about virtually every popular book about economics. But these are the books I was asked to review, and if anybody asked me whether they should read either one of them, I would reply by pointing that person to Krugman and Noah’s online writing instead. Because that is much more digestible, and even fun to read.

I didn’t have a huge amount of space to go into the substance of the books’ arguments, but on Twitter today, Noah took exception to my characterization of how he presents the importance of the Taft-Hartley Act of 1947. In the review, I say that after giving us chapter and verse on the act, Noah’s ultimate point is that the act wasn’t actually all that important in terms of rising inequality. Noah responds that I’m wrong: Taft-Hartley was very important, he says, and that was actually “kind of the chapter’s point”.

Now I’m not one of those post-modernists who thinks that an author has no privileged access to his own work, and that my reading of the chapter is just as valid as Noah’s. If he says that he was saying that Taft-Hartley was very important, then I’ll take him at his word. But that’s not the impression I got, either from the chapter in particular or from the book as a whole.

The chapter in question is called “The Fall of Detroit”, and covers the decline of unions in the US. In terms of the number of union members, the decline began in 1979; in terms of the percentage of the population, it began much earlier, in 1954. And it’s possible to blame some amount of the rise in inequality to the decline in unions. Noah cites Berkeley’s David Card, who reckons that somewhere between 15% and 20% of the growth in male income inequality between 1973 and 1993 was attributable to the decline of unions, while among women the impact was minimal. Noah, of course, is talking about a divergence in inequality from 1979 to the present day, most of which took place after 1993, and it’s fair to assume that if unions had already pretty much declined by 1993, the effect of their further decline over the past 20 years is slim. So if you’re looking for causes of the Great Divergence, I don’t think you can reasonably make a case that the decline of unions is responsible for more than 10% of it.

And if the decline of unions as a secular trend is a minor but still important part of the reason why inequality has increased so much, the next question is the degree to which the Taft-Hartley Act was responsible for that decline.

Noah’s argument here is subtle. It has to be, because the 1950 Treaty of Detroit, a high point in the history of unions, came three years after the Taft-Hartley Act became law. Still, writes Noah, Taft-Hartley was a kind of ticking time-bomb:

Another reason unions fell fast and hard was that the Treaty of Detroit, formidable though it was when constructed in 1950, lay atop the fault line of an antilabor law whose passage big labor had been unable to prevent three years earlier. If the 1935 Wagner Act was labor’s Magna Carta, the Taft-Hartley Act was its Little Bighorn.

Are the unions represented by George Custer and the 7th Cavalry Regiment, in this metaphor? If so, it doesn’t really work: Custer died at Little Bighorn, while the unions’ best days — like the Treaty of Detroit — were still ahead of them when Taft-Hartley was passed. But in any case, Noah’s basically saying that Taft-Hartley allowed corporate management to undercut labor in the decades ahead, and that as a result the Treaty of Detroit was much less of a victory than it might otherwise have been:

The momentum enjoyed by the labor movement and the remarkable job-creating postwar prosperity that would emerge within a few years (and on which big labor would come to depend) obscured for a couple of decades what a powerful weapon Taft-Hartley placed in management’s hands. “After ten years of experience” with the law, the University of Buffalo economist Joseph Shister wrote in 1958, “this controversial piece of legislation can be viewed with considerably less emotion.” Shister concluded that while the law had made it somewhat more difficult for unions to organize, the power relationship between management and labor was essentially unchanged. That judgment was correct for 1958, but it didn’t remain so.

So Noah has to make the case that because of Taft-Hartley, a lot of bad things happened to unions in the decades after 1958 that wouldn’t have happened otherwise. And frankly it’s a hard case to make, because elsewhere in the chapter Noah quite convincingly explains that bad things were going to happen to unions no matter what:

Management and labor were more adversarial in the United States than elsewhere. Mechanisms for compromise, either public or private, were few, and there was little tradition of joint economic stewardship. The resultant conflict made old-line industrial unions appear, to much of the public, maddeningly intransigent as the Rust Belt fell into steep decline. Some unions, like the Teamsters, were blatantly corrupt, with extensive ties to organized crime. That didn’t help labor’s image either.

But an underlying reason for labor intransigence was that Reuther was never able to build on the Treaty of Detroit sufficiently to establish a partnership between labor, management, and government comparable to what western Europe achieved after the war. American management wouldn’t allow it. It was too socialistic, too impertinent. When a corporate leader believed that Reuther had an excellent idea about how to run his business, he still felt compelled to reject it, on principle.

That, it seems to me — and this was certainly the impression I got from reading Noah’s book — is the real heart of the reason why America’s unions declined. In Germany, union representatives were invited onto corporate boards; in the US, they were treated as the enemy.

More broadly, in the book, Noah explains that the ovewhelming majority of the rise in inequality cannot be attributed to any one cause, like the decline of unions: it’s a much broader and subtler political phenomenon. Here’s where Noah completely convinces me:

Economists and political scientists previously resisted blaming the Great Divergence on government mainly because it didn’t show up when they looked at the changing distribution of income taxes…

But recently a few prominent economists and political scientists have suggested looking at the question differently. Rather than consider only taxes and benefits, they recommend looking at what MIT’s Frank Levy and Peter Temin call “institutions and norms.” It’s a vague phrase, but in practice what it mostly means is “stuff the government did, or didn’t do, in more ways than we can count.” In Levy and Temin’s view, the Great Divergence was the product of “a shift in the political environment.” Great income inequality, they wrote, would be impossible to achieve “without government intervention and changes in private sector behavior.” The two were mutually reinforcing.

In his conclusion, Noah writes that:

Today it can feel as though we live in a society that’s the precise opposite of Rawls’s ideal. The first principle isn’t economic equality; it’s economic inequality. Any effort to minimize income differences is held politically suspect, an intrusion on individual liberty.

I think he’s right about this. If Taft-Hartley hadn’t already passed in 1947, it would have passed some time later: as we just learned again in Wisconsin, the anti-union sentiment that allowed Taft-Hartley to get a two-thirds majority in both houses, enough to override Harry Truman’s veto, never really went away. Taft-Hartley is a symptom of a much broader syndrome; it’s not a significant direct cause of today’s inequality. And that’s why repealing Taft-Hartley would be so ineffectual as a weapon in the war against inequality. The real problems are deeply embedded in American society, rather than being enshrined in some labor-relations law. Do what you like to Taft-Hartley: the rich still run this country. And will continue to extract as much as they can in the way of rents.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

Unions stopped organizing for about 30 years after the Treaty. By the time the woke up, they were under DOJ investigation and held in low esteem by most non-union workers.

But public sector unions continued the tradition, and that kept overall income inequality from going to the redline right away.

Of course, it exacerbated the union/non-union divide by making it also a government/non-government divide.

The public sector boys were not averse to the kinds of perks and associations that animated LIUNA and the Teamsters, either. Consider the late Dom Badolato: cfm?action=article&articleID=17661bcf-8e b7-4b82-8e36-513f33d04dd3

He had a building named for him by the early 1990s (when he was still Council 4 Ex. Dir)–and a job with at least three different paychecks rolling in every other week. Plus a pension as well, I believe.

His memory is, of course, revered.

Posted by Eericsonjr | Report as abusive

Ummm… felix? German unions weren’t “invited” into the boardroom by friendly capitalists. A Social Democratic government passed a law requiring that employee representatives get half the seats on the supervisory board.

About Taft-Hartley: if you don’t think that increasing union density in the US would reduce inequality (which seems to me to be a very peculiar thing to believe), ok. But if you do think it would reduce inequality, are you saying that repealing TH wouldn’t boost density? Without TH, you would have card-check recognition. You would have legal sit-down strikes. You would have legal secondary boycotts. Without TH there would be no right to work states. So I find it very hard to believe that repealing TH would not lead to a substantial increase in union density. I also find it hard to believe that repealing TH would not lead to a substantial decline in inequality, since the growing gap between production worker compensation and productivity seems, inevitably, to be a the primary soucre of increasing inequality.

Posted by richclayton | Report as abusive

I’m surprised Noah didn’t cite Bruce Western’s and Jake Rosenfeld’s “Unions, Norms, and the Rise in U.S. Wage Inequality” as it gives more support to his argument. They find that 1/5 to 1/3 of the increase in wage inequality results from the decline of unions 1973-2007.

Posted by rowedj | Report as abusive

I think it’s not legislation nearly as much as it is that much of the world is catching up in skills, and you would expect wages to become more equal as skills do. Germany is at least partly a bogus comparison, because it’s clear that what has happened in the US is getting ready to happen in Europe.

As with an equalizing function, there are parts (financial services, executive management in general) that are not yet equalizing. Give it a few decades; I bet they will too. The trend is neither smooth nor fair. While society doesn’t think well in longer timeframes, that’s usually how dramatic shifts occur.

Posted by Curmudgeon | Report as abusive

I’m with Felix on this one: I don’t think inequality is something you can easily fix with changes on laws. Laws are only effective when they don’t fight against other strong incentives. People that think that laws would make a big difference probably don’t see clearly the forces that are maintaining inequality.

Inequality is usually the result of very competitive societies. When people compete, there are always winners and losers, and the winners usually take over at least part of the wealth of the losers. This creates more inequality. But instead of blaming excessive competition, losers are told to stand up again and compete harder than ever for the hope of winning this time.

To reduce inequality in America, for starters you would probably have to reduce the cult of competition, and recognize that while a bit of competition can be very beneficial, an excess leads to massive inequality, which leads to general unhappiness. But collaboration is often seen as “socialist”, which for many is about the same thing as evil.

But a change of attitude isn’t going to happen in the middle of a recession/depression, because when almost everybody are worrying about their own situation the urge to compete is stronger than ever. Even traditionally less competitive countries are getting more competitive now.

Posted by Doly | Report as abusive

@Doly @ 8:23 pm: “I’m with Felix on this one: I don’t think inequality is something you can easily fix with changes on laws. Laws are only effective when they don’t fight against other strong incentives. People that think that laws would make a big difference probably don’t see clearly the forces that are maintaining inequality.”

Nice try in justifying the 1%’s control of wealth and upwards redistribution. Here’s a law who’s ‘deregulation’ has had an enormous impact on this: USURY 528/is_1_53/ai_n25019601/

Posted by crocodilechuck | Report as abusive

Inequality can be fixed by laws, mainly the tax laws, but it probably won’t happen soon. It’s not that we need laws that say companies must pay people a certain percentage of their revenue or profits, but there are more effective and efficient ways to accomplish this goal, if the nation really cares about such things (which it should).

Corporations are virtual machines that were originally designed to act as vehicles for ventures that are too large for individuals to finance. However, over time and under tremendous social pressures, they have been transformed into a tool for executives to extract wealth from society. The corporation, founded for a specific purpose, to sell a product or service, instead becomes optimized to create profits, and specifically, after-tax profits. For the most part, they have excelled at this task. And as part of their design, they have manipulated governments to minimize their taxes, because what matters to the executives most is after-tax income. Not whether or not their employees earn enough to be part of their ecosystem that buys their products. Nor if their business can be sustained without extracting rents from the communities they help comprise. These details are insignificant to their goals of maximizing profits, and more importantly, executive compensation.

The tax laws not only enable this kind of attitude, but reinforce it. There are no penalties for hoarding profits, which effectively removes capital from circulation. There are penalties, though, in the form of double taxation, for distributing profits to shareholders, profits that otherwise might be re-invested. There are incentives for minimizing spending on research and development, building modernization, employee training and health care, and all those things that create jobs.

So here I go again saying we need a tax system that strongly discourages hoarding and rewards investment. Money that is kept out of circulation effectively reduces income for some other part of society. When we have a tax system that gives businesses no choice but to invest or distribute profits, then there will be more of those profits invested in things that create jobs, or pay people more, and the income inequality will start to disappear.

We can’t create a precise model for the economy so that income is distributed more broadly, but we can devise a system that gives the economy less options for preventing a wide distribution. And that is something legislation can accomplish.

Posted by KenG_CA | Report as abusive

I’m with richclayton. Your fourth paragraph makes extremely little sense to me; the absence of collective bargaining is something with a constant effect on your bargaining position, making any and all attempts to enlarge the divergence bigger easier.
Furthermore, where did the public unions go in this story of yours? (And how do you quantify the effect their example pay set during some period or other?)

Posted by Foppe | Report as abusive

In the NYT today, I review the new books from Paul Krugman and Tim Noah. Short version: these are both really smart people, whom you should pay attention to.

Posted by Anonymous | Report as abusive

Please change the article title. It should be “The Great Diveregence – On union busting legislation and its effects on US inquality”.

Posted by krimsonpage | Report as abusive

Wow, what a terrible review. It amounts to:

“Krugman and Noah’s books don’t answer some question they didn’t seek to answer, but that I want answered, so they stink.”

More Salmonian arrogant idiocy.

Posted by EconomistDuNort | Report as abusive