The decline of the U.S. labor share of national income

That’s the new paper by Elsby, Jobijn, and Sahin, presented at Brookings earlier in the week.  It’s less pathbreaking than some people are suggesting, but it is absolutely on the mark.  The main finding of significance is that competition from cheap imports is a major source driving wage declines in the United States and shifting income toward owners of capital.  The full abstract is here, with other points of note:

Over the past quarter century, labor’s share of income in the United States has trended downwards, reaching its lowest level in the postwar period after the Great Recession. Detailed examination of the magnitude, determinants and implications of this decline delivers five conclusions. First, around one third of the decline in the published labor share is an artifact of a progressive understatement of the labor income of the self-employed underlying the headline measure. Second, movements in labor’s share are not a feature solely of recent U.S. history: The relative stability of the aggregate labor share prior to the 1980s in fact veiled substantial, though offsetting, movements in labor shares within industries. By contrast, the recent decline has been dominated by trade and manufacturing sectors. Third, U.S. data provide limited support for neoclassical explanations based on the substitution of capital for (unskilled) labor to exploit technical change embodied in new capital goods. Fourth, institutional explanations based on the decline in unionization also receive weak support. Finally, we provide evidence that highlights the offshoring of the labor-intensive component of the U.S. supply chain as a leading potential explanation of the decline in the U.S. labor share over the past 25 years.

As I reported two weeks ago, Autor, Dorn, and Hanson already found similar results.

There is an entire chapter in Average is Over suggesting that trade effects on U.S. wages, in the negative direction, are stronger than many economists think, through factor price arbitrage, and that the topic deserves further investigation.  But it turns out my discussion did not go far enough in the direction of attributing observed wage changes to trade, and because of this paper, and because of Autor, Dorn, and Hanson, I hereby revise my views accordingly.

Please note of course that this trade is still output-expanding and welfare-improving by traditional criteria, at the global scale.  It simply has within-nation distribution effects which not everyone likes, though global inequality falls.

Arnold Kling, Michael Mandel, and Rob Atkinson, among others, have been ahead of the consensus on this question.


And so the great intra-libertarian contest between the cosmopolitans and conservatives continues...

It's interesting that this conclusion, properly nuanced, is basically identical to that which has been held by the man in the street (the guy who used to work in the labor intensive component of the US supply chain) for at least a generation. It is also a position that, until recently, economists routinely ridiculed...

Click the link. It goes:

Looking at other pieces of business sector income over those two time periods, they find that the decline of the labor share does not reflect an increase in corporate profit rates but rather an increase in the share of income paid for the use of structures and equipment. ...

The headline measure overstates the decline of the labor share by one-third because of the way self-employment income is estimated;
The declining labor share is not merely a recent phenomenon -- although the most recent decline has been dominated by changes in the trade and manufacturing sectors, even prior to 1980, there have been substantial, though offsetting, movements in labor shares within industries;
There is limited evidence that the share is dropping due to the substitution of capital for (unskilled) labor to exploit technical change embodied in new capital goods; and
The decline in the percent of unionized workers in the workplace is not, in fact, a major cause of the decline of the labor share.

which goes well beyond 'nuanced' in its differences from the man-on-the-street narrative.

I was referring to this conclusion, as, basically, was Tyler:

"Finally, we provide evidence that highlights the offshoring of the labor-intensive component of the U.S. supply chain as a leading potential explanation of the decline in the U.S. labor share over the past 25 years."

If ten or twenty years ago, a worker in the labor-intensive component of the U.S. supply chain had the temerity to express this concern to an economist, he or she would almost certainly been ridiculed as an economic ignoramus...

It's almost as if those beefy guys in windbreakers down at the union local were right about certain things that economists turned out to be wrong about.

Of course, that can't possibly be true.

Alternatively, it turns out Pat Buchanan was right about a lot of things.

So you rather have money go to an unskilled American rather than an unskilled Pakistani, black man, or Asian? Just to keep the American happy? The theory being he can cause mayhem if he's not paid off? Sounds like the racist SS SS we all know and loath (love?), LOL. Like TC's last sentence says: "Please note of course that this trade is still output-expanding and welfare-improving by traditional criteria, at the global scale." I agree, and also believe we should have 100% open borders, no restrictions whatsoever. So does AlexT I bet. Trillion dollar bills lying on the sidewalk with open borders.

Ray - True open borders are going to require open elections. Either that, or a further decrease in the strength of representative government in this country.

Should all the world's population be able to vote in the elections of the U.S.? I'd imagine they'd vote for pure open borders.

Would the reverse also be true? Should Americans be able to vote in British, French, Brazilian, Indian, and Mexican elections?

Ray Lopez: Absolutely. Those are the folks in my own backyard. They're a part of my polity's social contract and I may have to rely on them for disaster relief, mutual defense, the civil order, or even just a minimally polite society. You should go to Pakistan or the Congo and tell them why they should care as much about unskilled Americans as unskilled Pakistanis or Congolese and see what they tell you. If you think the American collective is immoral, then scrap it but the next day you'll see people starting a replacement collective faster than you can say 'blood is thicker than water.'

Also, this spergasm is hilarious: " I agree, and also believe we should have 100% open borders, no restrictions whatsoever. So does AlexT I bet. Trillion dollar bills lying on the sidewalk with open borders."

"No restrictions whatsoever" - really?

Net tax consumers, people with communicable diseases? Al Qaida? Idi Amin? Shining Path guerillas? The Zeta cartel? Han Chinese immigration into Tibet? Spanish immigration into Meso-America? Jewish immigration into Palestine? Mohammedan immigration into the Levant?

Idle resources are lying on the sidewalk, so to speak, all over the globe. Capital remains rather particular about where it will locate and who it will hire. Are any libertarian academics snapping up cheap land in the Rust Belt instead of bribing the Honduran government to give prime tropical coastland to their rich buddies? Is Silicon Valley sending buses to Detroit to bring back all those potential software engineers?

I assume as well the words 'open borders' are carefully chosen, because 'no borders' means that people get to draw their own.

Potential software engineers? Nobody hires potential software engineers. You hire actual engineers, which are probably local. For people that are actually qualified, the market is actually better in the midwest than in the Bay.

And you can't expect a company to train anybody nowadays. In most of those jobs, an untrained employee provides negative value for months, if not years. After being trained, they are just as valuable in a company that isn't training employees, and is just spending their resources on trained people only, therefore being capable of providing higher salaries.

Without contracts that make sure you can't bail, or social customs that make leaving something that is frowned upon, most training will come from the worker's pocket.

Surprise of the year: the word "immigration" doesn't appear in the article.


When I read the phrase "competition from cheap imports" I thought for a second that they were referring to the supply of labor.

Outsourcing jobs and insourcing jobs are fairly similar concepts, but it's okay to think about the former's contributions to inequality, but a comically high proportion of the economics papers poor Tyler cites studiously ignore the very existence of the massive insourcing of jobs that has been going on for two generations.

As the Wizard of Oz said to Richard Pryor's wife: "Pay no attention to your lying eyes!"

How do you reconcile your view that 'Mexicans are taking our jobs!' with your view that 'Mexican-Americans can't carry Italian-Americans' jocks'?

please, wogs are wogs to Sailer. Only pure anglo saxons losing out on these jobs are the victims.

Most Americans are not in competition with illegal immigrants That's true only at the bottom of the income scale. That doesn't mean it's unimportant, but at higher levels we are affected far more outsourcing and automation than by guys sneaking across the border.

According to this logic, the rich should continue to hammer down the standard of living of lower income American citizens by insourcing jobs because, hey, the victims are lower income, so let's not even mention this problem in papers that purport to be comprehensive. In contrast, outsourcing jobs hurts less vulnerable Americans, so economists darn well worry about that!

I agree, even though I am a strident supporter of immigration.

Tyler - Would you do a post on how this information affects your priors on immigration and low-income American workers?

Although I believe immigration on the whole is a huge benefit to America and the world, I am skeptical of the Matt Yglesias argument that low-skilled immigrants are actually complimentary to low-skilled American workers.

Those of us who are pro-immigration should grapple with the effect on low-wage American workers, instead of just sweeping it under the rug.

Is it not great when some True Believer acknowledges that his position is completely immoral? When he acknowledges that it is destructive for his coethnics but, nevertheless, he keeps his faith because he feels so intellectually sophisticated by holding that ideology?

Memo to fanatics: no, your mechanistic system of beliefs is not improving your social status by making you look as a sophisticated and high-minded bon-vivant. I know you cannot believe it, but trust me: it's making you look like a starry-eyed fanatic.

Massive migration from unrelated peoples is planned ethnocide. Period. Live with that, traitors. Live with your "internal redistribution of resources", immoral bastards.

So Ross Perot was right?

I suppose, though the number of jobs outsourced to Mexico pales in comparison to those outsourced to China, India, et al. So my guess is that the number of jobs outsourced to Mexico would have been only slightly lower without NAFTA.

Perhaps the Mexicans came to the jobs.

There you go again: Noticing Stuff. You should know that no respectable economist notices the insourcing of jobs.

Perot was wrong about Mexico and NAFTA. We had a jobs boom after NAFTA went into effect (No, I am not saying "because of NAFTA"). Mexico's economy is well-enough integrated into ours that we didn't suffer any real losses from NAFTA. Perot failed to see the tsunami coming from China and India.

(In the parallel universe of Libertania)

Ha. I posted above before seeing this. My version is "Pat Buchanan was right." A prophet is not without honor, except in his hometown and among his relatives and in his own household.

I always question the effects of supply and demand. For example if we increase the supply of labor does that mean the price of labor (wages) decreases? Not so, I am told, because if the supply of labor was constrained by thin labor markets then increasing labor will improve output and demand. But that isn't necessarily always true. And isn't increased trade effectively increasing labor markets?

The end of Communism in Russia and Eastern Europe combined with the opening of China to world trade and Republican deficit spending and a large US structural current account deficit lead to a massive increase in the world supply of labor. Consequently, labor became the cheap factor of production and capital was the scarce factor. So returns to labor fell and the returns to capital improved.

Not to mention a large increase via immigration in the number of workers to do the jobs in America that are difficult to outsource.

Whether it is technology or trade (trade is a kind of technology) the political response should be the same, use tax and expenditure policy to shift after tax income toward those who suffer from the change. That's what Pareto optimum is about. It would help, of course if we were taxing consumption progressively than income sort of progressively.

Yeah, let's get the central planners involved at the deepest levels, picking winners and losers and determining production and income distribution. That always seems to work out just fine.

And what do you believe the solution is to those who are suffering the most from the change? Or do you just not care?

One reasonable proposal is: Libertarianism in one country. Having American citizens compete against each other is one thing, having 7 billion other people compete all out against American citizens is another.

The whole project of Modernity is based on the most extreme individualism, predicated on a über-abstract concept of the individual human. That's why you cannot make a libtardarian understand that, yes, to destroy the life of millions of blood-related euroamericans is WRONG. He is just an abstract individual maximizing some mathematical function for the well-being of the whole planet, or universe, or several multiverses. He's so sophisticated. So cool.

The establishment wing of the GOP and the Clinton/Rubin wing of the Democrats both celebrate this as progress.

And I have been one of those ignorant fools ridiculed by the Very Smart People (VSPs) and the Very Rich People (VRPs) because I don't understand the "benefits" of trade.

Dozens of empty factories in a region is "anecdotal evidence" and therefore irrelevant to economists.

Sure miss the sight of the legions of happy auto workers filing into the Rambler plant in Kenosha for a day of bolting together those beautiful Ambassadors. Unfortunately, not enough drivers wanted to be behind the wheel of one but in a well-governed world we'd still be able to purchase a new one (including all the mandatory safety features and gasoline options) and those lower order laborers that you academics express so much concern for would still be employed there.

1. Things change

2. E.g., steelmakers produce a lot more with a lot fewer workers nowadays.

I knew that already.

>First, around one third of the decline in the published labor share is an artifact of a progressive understatement of the labor income of the self-employed underlying the headline measure.

Trying to figure this out. The headline measurement of labor income for workers on payroll is straightforward, just take the payroll reports. For self employed, the assumption is that the hourly income is the same as those on payroll, and the hours decreased from 1948 at 14% to 8.5%. The upward correction they make is mostly income due to rents and returns on capital.

Makes sense. All the business consultant types that focus on small business have for a long long time emphasized that the returns on capital are more worthy of focus than the returns on operations.

We are all bankers now.

Feels like "all inequality, all the time" here at MR lately. Oh well.

At least you gave a passing nod to a "global GINI" story of decreasing inequality, which hints at the real underlying issue- the same issue that has vexed history for all time, but one which is shrinking meaningfully- viz., the fact that we continue to inhabit a planet overwhelmingly populated by poor people.

“all inequality, all the time”

Average is over - you just haven't had your personally targetted algorithmic marketing system make you feel better about it yet, that's all.

Tyler notes the paper's most critical observation, but then breezes past it, "competition from cheap imports is a major source driving wage declines in the United States and shifting income toward OWNERS OF CAPITAL" (emphasis added). This isn't a story of certain (industrial) types of workers versus other (cognitive) workers, or even of workers in one part of the world versus another. Rather, it's the story of competition between those who can claim ownership--and can marshal the force of government to defend those claims--versus those who have nothing to sell but their labor power. That is to say, a class struggle.

Belief in Social Darwinism is a prior here I fear.

Definitely seems to counter the argument of ever-increasing Meritocracy that he makes without that prior.


“competition from cheap imports [OF LABOR] is a major source driving wage declines in the United States and shifting income toward OWNERS OF CAPITAL”

Again, how is any of this news?

A friend of mine, way back in 1990, was fond of saying "Third world skills equals third world wages in the 21st century." OK, he was a bit of a cold bastard, but the idea that the economics profession was blindsided by this or that this or that political decision brought us to the present state is extremely facile.

Brian, I would recommend reading the introduction of the paper: I think it spells out a little more the hypotheses tested in the paper ... and yes, your friend may have been on to something about global labor markets. I guess my quibble with the paper's findings is that outsourcing is not independent of technological (and institutional-tax-regs) changes or shifts in domestic demand (we've had some lately) and thus I am still not sure what is driving all this. Also note the paper throws water on TC's tech/labor complementarity thesis.

Good empirical economics, like this paper, is usually about documenting which anecdotes or theories hold water and which don't ... not about dreaming up new ones. I would discount all the talk in the post about novelty and who talked about this first ... it's an academic economist thing. What matters more is what's going on and how should the results affect our views on the economy and policy..and that's how the paper reads.

Unilateral tariff disarmament and de facto open borders were both political decisions.

well, where's the narrative drama in *that*?


Underneath your characteristically, comically apocalyptic portrayal, I discern an argument.

It's mostly a red herring- that's my point.

Now, after the posturing and the intellectualoid effete feint, negate that "unilateral tariff disarmament and de facto open borders were both political decisions". Negate that.

What about the effect of interest rates on labor's share of income? Low interest rates move markets up and since a substantial portion of executive pay is in the form of stock options and the like, they help top executives at the expense of other employees who do not own stock or are otherwise not in a position to benefit from the rising prices of financial assets.

Take a look at the graphs in

Please have a look at and the linked study

This CBO study is indeed subtle. Tim Taylor's concluging paragraph: "From all of this, here are a few take-away thoughts: The labor share of income in the U.S. economy is historically low. Some of this will self-correct as employment gradually rises again. Some of it is due to change in technology and globalization that have raised the return to capital and put downward pressure on the pay of substantial groups of workers. Some of the fall in labor's share of income is due, in a statistical sense, to less-discussed factors affecting capital income like a rise in recent decades in the value of living in a home that you own, and to the fact that capital investment in information technology has a shorter life and depreciates faster than past capital investment."

The complete set of "less-discussed" factors (more than the couple mentioned above) rarely make it to the discussion.

Is there proof that returns to capital have increased. I would think that record low interest rates would be an indicator that capital returns are poor. Otherwise the demand for loans to finance all these highly profitable capital assets should handily drive up interest rates. Perhaps "money printing" is to blame for low interest rates, but I would include uncertainty as a major factor in slow employment growth and therefore the falling labor share of income. With considerable uncertainty regarding the prospects on new business investments from shaky economies and fierce global competition, a prudent business would have a higher required rate of return before taking a risk.

Labors falling share started around 1980. What was there in Reaga onr Volckers policies that created such uncertainty in the 1980s?

Is uncertainty greater now than in the 1980s?

Yes this seems obvious.

But wouldn't it be the case that market forces and commoditization may in the future work to diminish the share of the capital owners (to the benefit of labor and consumers)?? Many industries deserve to see their profit margins squeezed and it may actually happen.

but of course it is not polite in mainstream circles to point out that mass immigration and diversity itself are both major sources of wage declines.

It will become slightly less impolite to think about the effects of immigration on inequality in America once some version of the Schumer-Rubio bill is signed by Obama, but until then: No Thinking about Immigration Allowed.

So the same person in India can't compete with you from across the globe, in a location with a lower cost of living? Lucky you. I'd rather have the same Indian competing from the apartment complex across the street.

Again pushing for the ethnocide...

For them, there's no political economy. Like if it were impossible to avoid the insourcing *and* the outsourcing. But the best part is their idealism (in a purely philosophical sens). In their minds there's no flesh-and-blood people, just abstract concepts, atomically interacting in a platonic world. Their humans have no reproductive interests. They're absolutely unable of thinking outside the libtardarian economic reductionism. Worse than useless.

Where are the studies reporting the rise in share of consumption by capital and the decline in share of consumption?

All indications are that labor has reduced consumption as labor share of production has declined, AND capital has reduced consumption as capital share of production has risen.

And as government share of production has fallen, government share of consumption has risen, and that is a trend that the Bush administration initiated in 2001 because for government to maintain its consumption below its share of production to pay off debt was considered damaging to the financial system that depends on the security of owning government debt incurred from over consumption.

I find it interesting that most economists are all one handed, either production handed, or consumption handed, so rather than marking time, economists are simply waving one hand, sometimes rather frantically.

This post is basically just hand waving.

Comments for this post are closed