Recent Declines in Labor’s Share of US Income

That is a new and important paper by Robert Z. Lawrence.  It is a little hard to excerpt, but the core messages are pretty simple:

1. Labor and capital are mostly complements.

2. The recent problems of labor are due to a lack of capital, not substitution of capital for labor.

If Lawrence is right — and he has plenty of data on his side — a lot of what you read about these topics is wrong, at least circa the status quo.  And the idea that we need stiffer taxes on capital income could be disastrously off base.  This paper is interesting throughout, yet I predict it will be largely ignored for its inconvenient nature.

For the pointer I thank Robin Hanson.


"This paper is interesting throughout, yet I predict it will be largely ignored for its inconvenient nature." If only there were some way to determine if Lawrence is right.

In Econ. you can almost always find a reasonable sounding paper to support mostly any viewpoint. Hardest thing is to know who to ignore.

One would think the economics profession had long ago determined the natures of labor and capital... and their relationship.
Apparently not. Basics are so mysterious.

Why would one think that? The truth of the matter is the "nature" of the things aren't even fixed, so how could their relationship be?

If so, then what's the point in studying something so ephemeral?


The questions, they press. So the answers, not so good, but the questions, they press.

You know how we each have an internal running monologue, a "Joycean stream of consciousness" that is constantly updating the Narrative of Me and My Place in The Cosmos.

Perhaps social sciences is a running monologue society has with itself, ever tentative, ever contingent, but something to do as the world turns.


Maybe, you are right!

Rahul, economics has figured out that one. You study something because someone is paying you to do it. economic metaphysics 101.

I am stuck at why the US would be a complete enough "system" for this relationship to be valid. Surely investment overseas, often by US investors, upsets the math.

And as an aside much more of my consumption is straight from China via eBay this year. Only my mailman knows for sure.

You do realize that the biggest reason items from eBay from China are cheaper is because they have massively subsidized airmail prices.

You can get a 32 bit microcontroller board with embedded WiFi for $3. (ESP8266)

That is not just a shipping advantage.

Subsidized Yuan?

Open source intellectual property, low labor rates, high business connectivity, huge economies of scale, and Moore's Law. (This chip we play with as a stand-alone Internet of Things processor was built "as" a low cost WiFi dongle for mass market.)

Who subsidizes? Ebay or the Government? Just curious.

I just got a $15 Xaiomi Mi Band in the mail yesterday (vs a $80 Fitbit). I decided to look at the package to see if postage was specified. Strangely, even though I bought from Hong Kong, and there is much Chinese writing on the package, it is "Swiss Post." The return address is in Zurich. No idea what the deal is there. Does the Swiss Post Office offer cheapest shipping from Hong Kong?

Well, we might be looking at an epoch-making paper here, one that upends our understanding of the world in fundamental ways. They do come along every so often, and no one rings a bell to let you know which ones they are. So who knows?

But one thing we can be sure of--if this paper doesn't upend our understanding and rewrite the conventional understanding of labor and capital, it must be because of its "inconvenience."

How long is an epoch in Econ? Till the next recession?

this is too inside baseball for me but why can't both legs of 2. be true? can't there be a lack of capital and capital substitution for labor?

Point 1(complements) rules out the latter option.

Assume a heterogeneous market.

I don't see why it can't be both - part of capital can both be used to put more labor to work and also part can be used to chase around and bid up the price of certain assets (stocks, land, fancy schools, yachts, art, etc.).
There is plenty of additional work that could be done by people, and plenty of capital to put them to work - as always, the question is why is that capital not doing so? Perhaps because it is not profitable - there is no immediate investment return on some items, especially publicly financed public works that have no tolls. Even if advantageous to society, it may not be a good fit for private capital investment.

Does this mean that 1. If capital/worker increases, not only do wages go up but also wages/capital income go up, due to sigma<1? 2. Because of labor augmenting technology, capital/effective labor has fallen and hence so have wages, and 3. The income due to the augmenting part of effective labor, such as robots and patents, is counted as capital income, not labor income, which exaggerates this rise in capital share?

Marx was wrong. The conflict isn't between labor and the bourgeoisie, it's between the blue collars and the white collar bureaucrats, who can only be called labor in the sense that they must occasionally actually show up at their place of employment. Sitting in front of a computer from 9 to 5 may not qualify as "work". That's not the heart of the issue, however. Conditions are.

Lack of capital is due to a lack of investment.
Lack of investment is due to a lack of motivation.
Lack of motivation is due to a lack of demand.
Lack of demand is due to stagnant wages for labor.
Stagnant wages for labor is due to a decrease in labor share.
Decrease in labor share is due to an increase in capital share.

How did those capitalists manage to increase their share?

You missed monetary policy in that chain of causalities.

So where does it fit in?

Before the lack of demand.

Monetary Policy. Things like QE2. Increasing the supply of money, by giving more money to people, capitalists, generally, who already have a lot of money. Bailing out banks, but not borrowers. These activities would seem to further reduce the motivations of capitalists to invest in real capital. I'm not sure, but is that what you are getting at?

Lack of motivation could include looking at the list of pamphlets that have to be handed to an employee in California.

I'm surprised that they don't mandate every employee to have a lawyer paid for by the employer.

Damn, i wish we'd thought of that instead of the stupid pamphlet idea.

/s/ California lawyer

To be sure, obstacles to potential profit always enter into the calculations of a capitalist, when considering the profitability of an investment. But nothing is so discouraging to investment than the absence of a market. That is, an absence of demand, due to there being a limited supply of money, *in the market.*

The banks may be flush, the capitalist may be flush, but increasingly, labor is not.

Higher investment and faster real growth would certainly raise real wages at the bottom and middle, It hard to see how that would raise the labor share however desirable it would otherwise be. Agreed that we should shift from taxes on capital income. Unfortunately, current proposals to reduce capital taxation do not include progressive consumption taxation.

I'm not sure how "inconvenient" the finding is. The excerpts below seem strongly supportive of Summers' call (circa status quo?) for e.g. massive infrastructure investment:

"the cause of labor’s falling share recently is the weakness of investment...suggest[ing] that measures that boost investment and capital formation would lead to higher wages, raise labor’s share in income, and reduce income inequality" (47)

In fact, the author suggests a primary causal role of inadequate demand - perhaps inconvenient indeed:

"It is also paradoxical to think that enhancing labor’s productivity could reduce labor’s share in income, and reduce wage/ rental ratios... labor augmenting technical change, which at the margin encourages the use of labor rather than capital, is termed “capital-saving”; thus another implication of labor-augmenting technical change is that less capital is required to produce a given amount of output. This implies that if output is constrained by inadequate demand for example, investment would be weaker" (46)

Investment in productive capital (that which makes labor more productive) has declined, as owners of capital have increasingly chosen speculation (mostly in financial assets) over productive capital. Why? The rate of return on productive capital has been falling for years, while the returns from speculation have been spectacular; moreover, the Fed has signaled loudly that it will intervene to protect the gains realized from speculation in the event of a financial crisis triggered by over-zealous speculators. It isn't the taxation of capital that is the problem, it's the perceived lower rate of return form productive capital. An economy built on speculation is an economy with a declining future. Why has the rate of return on productive capital been falling? That's a good question and one readers won't see often, at least not here. The rate of return from productive capital has been falling for roughly the same period as inequality has been rising. Is that a mere coincidence, a correlation, or a cause? Cowen and like-minded people believe tax cuts for the wealthy (it's the wealthy who own capital) are the answer to every problem. But taxes can be cut, and cut, and cut, but as long as the rate of return from productive capital remains depressed, capital will seek a higher return elsewhere, and labor's share will continue to fall as will labor's productivity and the prospects for our future.

If I buy a financial asset from somebody, that person needs to do something with the cash -- invest in productive assets, or spend it on current consumption. Putting my income into financial assets doesn't consume any resources.
Investment in financial assets is not in competition with investment in physical capital.

Why can't they also invest in another financial asset?
The crash came in part because of the vast proliferation of exotic financial assets for investments (that turned out to be actually poor investments)

Who are they buying another financial asset from? What happens to that money the other guy gets for selling his financial asset?

You are on the threshold of "it's financial assets all the way down" territory, a kind of proof by contradiction.

I'm not so sure rayward is off the mark here. Financial speculation is almost entirely speculators to speculators trading with each other. They aren't ever going to invest in productive capital with those assets. Some of the profits do get siphoned off to pensioners and retirees, but obviously that is consumed not invested in plant.

I say all this as being a part of the financial industry. I am fine with markets and speculation, but I don't think it's breaking any true economic theories to see a fundamental disconnect between speculative traded capital and wealth creating capital.

rayward has been beating this 'speculator' drum for a while now. I'm ok with it if it means laying off his r>g drum, which was excruciating.

I don't even now what it means. To buy a stock is to buy a 'share' in something real. Does it matter if someone perceives himself as a 'trader', and 'investor', or a 'speculator'?

Rayward doesn't understand that the stock market is not like the bond market (which is almost purely financial) nor is it like the art market (which is purely speculative).

Perhaps you and rayward are hinting at the 'financialization' of the economy, which is a different argument that is basically wrong for different reasons from this odd 'speculation' grousing.

If this is true, exactly what sorts of capital in the American economy missing out on right now and exactly what sorts of jobs would these new investments create? I think being able to answer that question credibly would inspire more confidence in the underlying statistics.

Regarding the comment about taxes on capital, isn't this confusing two different definitions of the term "capital"? If we are talking about physical capital -- new buildings, production facilities and equipment -- it is hard to imagine that the tax system is what is discouraging companies from making additional investments. And even if you want to criticize the federal tax system, it is still the case that state and local governments will actively compete with each other in terms of offering tax breaks, subsidies and incentives for a company to start operations there. I've heard stories of even medium-sized employers getting serious face time with state governors at there mere suggestion that they might open up a factory or office and employee 500+ residents.

The debates over capital income seem much more focused on financial capital. As Tyler reminded us a few months ago, the average price-to-book ratio of companies on the S&P 500 has increased over time, meaning less and less of capital gains that people would realize on equity investments can be attributed to investment in physical capital. To put it in more concrete terms, when Facebook did its IPO, how much of the money it raised went into building new offices and data centers and investing in new equipment versus simply making Mark Zuckerberg rich? Are the returns Facebook's new investors have realized over time attributable to returns on its investments in physical capital or are they returns on intangibles?

Yes, I'd say the capital we lack is intangible capital, a kind that it is hard to create just by consuming less. It is easy to make more buildings and machines, and much harder to create new kinds of businesses and products.

US labor is in competition with the global poor for wages. No mystery here.

"And the idea that we need stiffer taxes on capital income could be disastrously off base."

LOL. Economists are worried the proles are getting restive.

US labor is in competition with the global poor for wages

Agreed. China's vaunted growth is mostly in coastal regions manufacturing goods to foreign designs that can then be stuffed in a container to be sold through WalMarts around the world.

Or stuffed into brand name items as components. Assembled in Vietnam or Malaysia.

Not enough capital? What do we do after interest rates at zero percent and no taxes on capital income?

Money isn't capital.

The money bidding long bonds down to 2% is sure capital and so is the money bidding the S&P 500 up to a P/E of 20 and all the cash sitting on the balance sheets of major corporations. The US, Germany and Japan are awash in capital.

American labor wages are stagnant because: (1) technical progress is essentially stagnant; (2) American workers have been brought into head on competition with Japanese, Chinese and Korean workers (the stagnation onset coincides with the surge in Japanese car manufacturing); (3) widespread transfer of American profits to tax havens (which strands the money and drags the economy)

#1 and 2 seem right. Is #3 true, though? Does money deposited in Swiss banks not find its way back into the economy in one way or another?

I think that the money does find its way into the US because an American company wants to hold US dollar denominated instruments. I expect we see those stranded profits passively invested in the Treasury market and in the excess bank reserves at the Fed. That is not where we want them - we want them spent or actively invested.

Rec3, it's atavistic to think of profits earned in overseas operations as American. A number of brand names - Apple, Colgate, e.g. -- have more sales overseas than in the US. And they have significant overseas manufacturing, distribution, etc. It's old-fashioned to treat the cash outcome of those sales and operations overseas as native to the US.

Last time I checked, taxes on capital income were between 20% (long-term) and 40% (short-term).

Those are the highest nominal marginal rates. With accelerated depreciation, 401(k) and IRA plans, pension trusts, charitable foundations, university endowments, tax exempt bonds, deferral of capital gains taxes until disposition etc the effect average rate of tax on capital income is much lower.

The U.S. allowed it's capital to be sent overseas to fund development in other economies, raising wages on foreign shores. Ricardo only works if your capital stays behind. Fed manipulation of interest rates deters capital formation (why save when you earn 0%) and encourages malinvestment in housing. Blowing trillions on wars didn't help. Government spending down to the local level is now mainly for labor, including health benefits and pensions. The seed corn is being eaten, or leveraged.

The real interest rate is going to spike far above expectations when the market realizes capital has been depleted and the central banks can no longer suppress rates. Here's a non gated link to the paper.

The taxes on capital should be zero, but U.S. policy should encourage capital formation and investment in the U.S. The approval of LNG terminals will determine how many billions of USD are invested in chemical factories in the US, with high wage jobs for Americans, versus chemical factories overseas. The U.S. ho industrial policy, has no care for it's citizens, so it doesn't even ask these questions.

Finally, when you have a capital deficit, you don't try to increase the labor pool and certainly not flood it with low skilled workers.

Foreign investment in the US is huge. Hence Bernanke's global savings glut hypothesis.

Actually... the US didn't just *allow* capital to be sent overseas, it incentivized it by being one of the only countries on earth to tax repatriated earnings.

If a company earns $1 in profit overseas, they can either invest that $1 overseas, or invest $0.65 in the US (post the 35% tax hit). Its kind of rough to justify investing that capital in the US when you have to make almost 50% return on your repatriated capital *just to get back to even*.

If a company earns $1 in profit overseas,

There is a huge problem right now with companies pretending that a $1 of profit was earned in Ireland or some other tax haven when in fact it was earned in the United States or Germany or France or Japan etc. Most of the stranded profits waiting for a repatriation holiday is there as a result of tax avoidance schemes that should be illegal.

"The real interest rate is going to spike far above expectations..."

In keeping with Cowen's third law:

You're on to it "8."

Like you said above: "Money isn't capital." Money is a token of *demand,* as I argue over at

Over 50,000 factories in the US have closed since 2001. Over 15% of our manufacturing capacity. That is not evidence of a growing economy. Yet we are told the economy *is* growing, even while our infrastructure is also declining,

So how much of our 'growing GDP' is due to increasing 'investment' in manipulating and capturing demand, and not increasing investment in production?

"Over 50,000 factories in the US have closed since 2001. Over 15% of our manufacturing capacity" - on the other hand US manufacturing output is at an all time high and the share of manufacturing as percentage of GDP remains constant.

Fake crisis's in manufacturing bases are one of the oldest tricks in the protectionist books. Try to think up some new problems please.

Would we expect the same wage/capital relationship for any State in the Union? For any County? Seems to me that if it does not hold at any scale, it isn't a rule.

Or put differently, dramatic global political change drowns any hope of "all else being equal" economics.

I just bought a 3d printer straight from Communist China. Who would have guessed either of those things 30 years ago?

With 10 Years a little over 2% and Germany & Japan under 2%, the developed world does not appear to be short potential capital. So why don't we have more investment? I wonder if business investment is becoming very productive and has high diminishing returns. Aren't IT investment dollars dropping the last ten years? But it would be hard to say the economy is getting less for the IT spend.

We have the "Unicorn" phenomenon, literally billions chasing thin business ideas. That would say to me that the savings glut is not distributed evenly. It is lumpy, with most in low yielding bonds, but some in desperate searches for mythical animals. (Mobile Internet ‘unicorn’ companies now valued at $575B worldwide)

Those low rates are government bonds. That money disappears down a multi colored hole with no productive return, and in many cases spent purposely to limit economic growth.

An economy where everyone has a car replaced when worn out, a house that is perfectly adequate and maintained, meals on time, kids in school, with cops keeping the peace and various bureaucrats running around making sure everything is in order, has 8% unemployment, high teens or higher youth unemployment and growth about equal to population growth, but in reality flat everywhere except a few hot spots.

What creates growth is something like the fracking revolution, or the high tech areas sprinkled throughout the country. Someone invests in a technology that make something new or something cheaper or more available, creating a ripple effect throughout the economy as others invest to take advantage of the change.

It is easier, cheaper and more profitable to do those things elsewhere. The investment money isn't used to build factories to fill foreign demand. It is to build factories in foreign countries to meed foreign demand.

@ 1

What is now the function of "labor" (of what kinds) in the income that is generated within the U S economy?

Is "labor," without differentiations and functional identifications, a useful, or distracting, category for economic investigations?

2. The recent problems of labor are due to a lack of capital, not substitution of capital for labor.

So the capital to labor ratio is lower? Presumably outsourcing and immigration have increased the labor supply and subsequently lowered the capital to labor ratio?

And the idea that we need stiffer taxes on capital income could be disastrously off base.

Why would this be the case? Employers would still have an incentive to increase profits, which requires increasing labor productivity, which requires more and better capital goods, which increases demand for more and better capital goods, etc.

"If Lawrence is right — and he has plenty of data on his side — a lot of what you read about these topics is wrong, at least circa the status quo. And the idea that we need stiffer taxes on capital income could be disastrously off base."

Oh lookee here a brand new economics paper that threatens to overturn the previous concensus with political implications I would support. Does it strike anyone else that maybe an academic displicine that repeatedly comes to confident conclusions that are just as confidently overturned by the next big paper that implies the opposite political prescriptions while at the same time declaring itself to be separate from politics is manned by some not-very-serious people?

"It shows that despite a rise in measured capital-labor ratios, labor-augmenting technical change in the US has been sufficiently rapid that effective capital-labor ratios have actually fallen in the sectors and industries that account for the largest portion of the declining labor share in income since 1980"

This is not a different way of saying "labour share of the income is decreasing because of labour-saving technological change?"

Labor is paid its marginal product.

How do you raise the marginal product of labor?

By making more capital available per worker.

Shifting a worker from a manufacturing job to restocking shelves at
Walmart lowers the capital per worker.

But the growth in the gross capital per workers has been on a downward slope since 1980. Moreover investment by those subject to the individual income tax has fallen even more contrary to the received wisdom.

Moreover, because of increased use of short lived IT equipment the growth of the net capital per workers has
slowed even more in the US.

The data to show these trends is freely available at the BEA.

This rather than Cowans' "low lying fruit" is the real reason for US stagnation.

"Shifting a worker from a manufacturing job to restocking shelves at Walmart lowers the capital per worker."

Only if the factory shuts down. If the factory just produces the same amount without the worker, capital per worker has increased.

At some point robots may stock the shelves at Wal-Mart more efficiently than humans. Prices will fall, there will be fewer employees, and even more capital per worker.

I'm a broken record on this, but it's fascinating to me how there is a seemingly unanimous feeling that there is a mystery to be solved regarding a savings glut and a lack of investment. According to the BEA. real housing consumption has been declining for 20 years. The housing "bubble" was partly a price reaction to constricted supply. Then, after 2006, we added constricted demand to the problem. Now nobody can buy a house either. So for nearly a decade, we don't let builders build where there is housing demand and a good chunk of potential buyers can't get credit. Housing would normally be a huge destination for capital. There is demand for housing consumption. Rent inflation has been high for 20 years, and continues to be high. Households are bidding up the stagnant supply of homes.

It's no mystery why capital can't find enough outlets. These constraints are lowering real incomes (through supply-side rent inflation), especially among low income renters in the major metro areas. Interest rates and real incomes will rise when (if) we allow building again. But, considering that many people consider today's (!) housing market to be a "bubble" that we need macroprudential regulation to pop, the worst could be yet to come.

No one can buy a house? Doesn't look like it to me. Look again. Its called the Boomer effect peaking in the mid-90's. Really a bad bad post. You don't get the demographic parts at all.

Granted it was an overstatement, but I was referring to the mortgage market and new building, which are dead. Of all the things I said, that shouldn't be controversial.

Falling demand would not be associated with persistently high rent inflation.

New construction is hopping around here, meaning Austin, Texas. Dead building markets are a regional thing, not a county-wide thing.

Well, you're right. Texas isn't part of the problem.

City--------------------Population------2015 1Q Housing Permits---Units/Thousand pop
Austin------------------2 million---------5,364------------------2.7
San Fran-Oakland--6 million---------2,411------------------0.4
NYC,NorthNJ,LongIs--20 million----10,580----------------0.9
San Diego-------------3 million---------2,251-----------------0.8
LA------------------------13 million-------9,816-----------------0.8
Washington DC--------6 million-------4,861-----------------0.8
Boston-------------------4.5 million------3,094----------------0.7
Chicago------------------9.5 million------3,124---------------0.3

And, there is no lack of demand for housing in those other 5 metro areas. Does that rate of new units even make up for loss of old units?
Except for Chicago and Austin, rental vacancy rates in all these cities are 4.6% or below and rent inflation is running high.


Interest rates remain at rock bottom which suggests that the supply of capital (relative to demand) is not particularly constrained.

look at real interest rates. again, think for once.

Yeah. Same story there.

Damn, Cowen's third law is money!

Aren't real interest rates lower, even negative?

Nominal interest rates were low during the Great Depression too, even as all the banks were failing. Interest rates are driven by inflation expectations and investor confidence and may not speak to the tightness of credit conditions.

One problem today is that our monetary policy is looking too hard at inflation and not hard enough at NGDP, which tends to push growth lower than it could be.

So Japan is experiencing a boom, right?

So real interest rates are sky high, right?

We're in the middle of the largest glut of liquidity, i.e. dollars, money, capital, in global history- the most money sitting around waiting for someone else to do something worth stealing, investing in, etc, and here comes a hero telling us that the problem is a "lack of capital" causing labor's share of income to fall. Sure, it seems like this is insane, because what "labor's share of income falling actually, you know, MEANS is "an ever increasing net amount of capital and increasing net capital formation rate". I.E. the world's capital stock is growing faster, and faster, and faster. But the magic ponies aren't bringing a change in labor's income back towards neutral! In fact, it's going the exact wrong way?

When commenters point out that this is an asinine, self-refuting observation, Robin Hanson pops up to say "no guys what I really meant is *imaginary* capital. You know. Good ideas and stuff!". Yeah. Labor's declining share of global income is because nobody invents anything anymore! For no reason, this human constant has suddenly changed! And the solution, guys, is to eliminate taxes on capital!

Because, you know, with a larger global savings glut, people will start thinking of good ideas again! Or, wait. Maybe they'll stop being discouraged from all of these gol-darn government regulations!

Earth to self-deluding ideological fuckheads: the current owners of capital are supplementing their current labor stock with capital where they see it to be productive to do so already. It's really easy to do, because capital is basically free right now. Making it marginally even cheaper is literally trying to run twice as large a hose into the ocean and waiting for the volcano to somehow form.

glastnost, if you announce your confusion about the difference between capital and money to begin an angry rant about an ideological conspiracy against labor, you probably shouldn't be throwing around terms like "self-deluding ideological fuckheads".

"this human constant"

That's an astonishing claim. So the rate of invention is the same in Silicon Valley and among Amazonian tribes?

Two things seem to be primarily driving the confusion:

1) Low nominal interest are very often confused with an "easy" or "loose" investment environment.

2) The proportion of added value an investor can "skim off" is much lower than commonly believed -- the vast majority of those gains accrete to consumers relatively quickly, where they increase the value of labor to consumers. This is actually where nearly all the gains in labor compensation come from -- the raw value of a person's labor isn't much more than it was 100 years ago (humans haven't changed much), but we have ways to employ that labor so much more effectively that his living standards are far higher (machines and business processes have changed a lot).

The gains flows to consumers at a certain rate as patents fall and competitors arise. Over time, without new gains being created, the increase in the value of labor seems to be happening more slowly. Responses to this that decrease investment will exacerbate the problem.

Comments for this post are closed