Felix Salmon

Chart of the day: The long decline of labor

By Felix Salmon
September 26, 2012


This chart comes from Margaret Jacobson and Filippo Occhino at the Cleveland Fed, and it’s reasonably terrifying — yet another one of those charts where the trend is down and to the right, and where it’s only gotten worse since the end of the recession.

What you’re looking at here is the share of total national income which is accounted for by labor — a measure that includes wages, salaries, bonuses and things like pension and insurance benefits. Everything else is capital income: interest, dividends, capital gains. There are two ways of measuring this, which is why there are two lines; both of them are telling the same story.

The fascinating thing to me, here, is what has happened since the crisis. Over the past three years or so, wages and salaries have been rising steadily, while interest rates have been stuck at zero. It’s never been harder to make income from capital, while incomes for people with jobs have actually kept on rising. And unemployment, while still high, has been coming down.

Given all that, it would stand to reason that the share of national income going to labor should be rising, not falling. Labor incomes are going up, the number of employed people is going up, and income from savings is going down. And yet! It turns out that people with capital are so rich, and getting so much richer, that it’s not even close. All that belly-aching about the plight of savers on fixed incomes in a zero interest-rate environment? Well, you don’t see it in these numbers. Looking at this chart, if you were given the choice between having money and no job, or having a job but no money, it’s not obvious which one to go for.

Of course, as the Cleveland Fed paper shows, a lot of the story here is about rising inequality. But the more powerful, if less obvious, story, is just how entrenched capital income has become in the US economy. As recently as 2000, it was at levels more or less in line with the historical average. And then, something big happened. During the Great Moderation — when yields fell on all capital asset classes — capital income went up sharply. Then the crisis happened, a classic case of a dog not barking: you’d expect capital income to have fallen enormously, at least for a year or two, but it didn’t, it just stopped rising. Most recently, in the wake of the financial crisis, capital income has been soaring again.

There’s a big lesson here for anybody serious about fiscal policy, too. (Paul Ryan, I’m looking at you.) As the labor share of income goes down and the capital share of income goes up, the only way that we can stop tax revenues from plunging disastrously is to tax capital income at least as much as we tax labor income. By contrast, the Ryan plan proposes taxing capital income at zero — putting ever more of a burden on working Americans, while giving unearned income a massive tax break the rich really don’t need.

There are big global forces driving this chart, most importantly the way in which labor is becoming increasingly global and fungible. Labor income has been declining for a good 25 years, and the only substantial countertrend was the dot-com bubble. The trend is a bad one, and it’s getting worse. And while I don’t see any policies, on either side of the aisle, which really try to address it, the fact is that Republican policies seem explicitly designed to exacerbate it. Think of capital income as the money flowing to “job creators”, and the chart is very clear on that front.

11 comments so far | RSS Comments RSS

“What you’re looking at here is the share of total national income which is accounted for by labor — a measure that includes wages, salaries, bonuses and things like pension and insurance benefits. Everything else is capital income: interest, dividends, capital gains. There are two ways of measuring this, which is why there are two lines; both of them are telling the same story.”

Err, no. The labour share and the capital share do not combine to equal 100% of national income.

There are four parts. Labour, capital, mixed income (sole proprietors, the self employed) and taxes minus subsidies. If we look at the UK figures for the same time period we find that the labour share has fallen, yes. But the profit share is pretty static (at least after the disaster of the mid 70s). What has risen is the number, and therefore total income, of the self employed. And taxes minus subsidies has risen too (in the UK experience, largely driven by higher VAT rates).

I agree, the US experience is not the same as the UK one. But I wouldn’t be surprised to see that something similar has happened in the US.

With this particular paper the researchers have done something I’m not sure is sensible. They’ve simply assigned the tax/subsidy and mixed income parts of national income to the labour and capital shares in whatever proportion those are.

“The remaining categories, proprietors’ income and indirect taxes less subsidies, are partly labor income and partly capital income, in proportion to UL and UK, respectively. As a result, labor’s share is computed as the ratio of unambiguous labor income to the sum of unambiguous labor and capital income, i.e., UL/(UL+UK).”

And I’m really not sure that that’s a valid way of doing it. Not if the reason the labour share is falling is because of a rise in either mixed or tax/subsidy share.

Posted by TimWorstall | Report as abusive

Labor law reform, which admittedly is not getting much attention, would certainly be one policy, recently promoted by the Democrats, that would address the falling labor share of output. The other big ones would be an aggressive use of fiscal policy to get back to full employment (like the American Jobs Act and then some), and debt reliefe for underwater homeowners and students (idirect, but it would free up current income for spending on newly produced goods and services, which would increase employment & wages).

Posted by richclayton | Report as abusive

You are making a significant error. Nearly all the change in productivity comes from rising commodity costs. Look at the BLS visual essay on the productivity compensation gap. There is an extract of the relevant graph on my blog.

At any rate, rising commodity prices = rising wages for brazilian miners. When you put it like that it doesnt seem so scary.

The basic story is that integration of the BRICs has expanded the labour pool at the lower end, without producing the equivalent number of Harvard graduates. Thus the supply at the top end is relatively constricted. At some point the BRICs will start having their own top tier education establishments, and the skill premium will be reduced as supply increases.

Posted by phil_20686 | Report as abusive

“As recently as 2000, it was at levels more or less in line with the historical average. And then, something big happened.”

I believe it was a two step process: first unionization declined, as the delusions of free market economics took hold, in large part thanks to economists, aligned with both centrist Democrats and conservative Republicans. Then in 2000 China joined the WTO and the bottom fell out from under the working and lower middle classes. Employers took advantage of low cost platforms overseas to hammer their domestic workforce through threat of job loss into ever shrinking paychecks and benefits. It you look at the chart of manufacturing job losses it parallels the one you show here to an eery degree.

Posted by nyet | Report as abusive

Echoing Tim Worstall’s point but adding some U.S.-specific context, I would also think about the impact of tax policy on how income is characterized. I agree that labor’s share of income has decreased, but I’d also like to see researchers try to adjust for several factors that more than likely skew the comparison over time.

One big factor that comes to mind is whether pre-1986 and post-1986 numbers for household capital income are comparable. Pre-1986 there were a lot more shelters available to depress capital income, and higher tax rates encouraged doing so.

Similarly, it appears the authors state that only realized capital gains – as opposed to unrealized – are a component of capital income. Obviously people have been encouraged to harvest unrealized gains by lower personal capital gains rates, as well as by the expectation that those rates were (or are) going to increase at the end of 2010 and 2012. There are also plenty of ways that income from a solely-owned or closely-held business can be characterized as either labor income or capital income, and that can be sensitive to the tax advantages of one versus the other. Recall, for example, Newt Gingrich paying himself a relatively small salary from his S-Corp but characterizing most of his income as “capital” income (dividends), which did not impact income tax due but did reduce his Medicare payroll tax burden. This latter point dovetails nicely with Mr. Worstall’s comment about “mixed” income.

In short, I think it would behoove a group of researchers to dig into the differing components over time and make sure that they are being presented on a comparable basis. It would be quite an undertaking that would require looking at each component and taking into account factors such as the number of self-employed people.

Posted by realist50 | Report as abusive

There are other problems.

First, the graph isn’t adjusted for demographics. And one life cycle event is that people earn money from labor in their younger years, and save some amount of that. In their later years they cash in those savings (sell things they bought as investment) – so the ratio of “labor earnings” to “capital earnings” will change for a given cohort over time. The baby boomers are big cohort, and they’re entering the “cash out investments for retirement” phase.

Second, some large part of compensation for many groups, most famously software people, is in the form of distributions of capital. As in stock options. Which show as ordinary income when exercised, BUT if not all cashed out, create shares of stock that the employee will sell for hopeful gains later. The net effect of this is to convert some part of the “labor wages” that person would have gotten into “capital returns” even though all of the money was in effect compensation for their work.

Third, while the shape of the graph shows a trend (but see comment on validty of numbers in comments above), the SIZE is misleading – the graph only spans 58% to 71%, while looking like it might span 0% to 100%. This is a classic trick from marketing 101 – for shame!

Posted by BryanWillman | Report as abusive

Given the evisceration of labor since about 1980, how do people increase their wages? I am not talking about the exceptionally talented — sports figures, people with advanced degrees or rare skills, or high-profile entertainment stars — but the average, run-of-the-mill people who clerk at grocery stores or come to fix a leaky pipe. What leverage do they have over their employers to claim their share of their much-more-productive-work-compared-to-19 80?

Posted by mlnberger | Report as abusive

Felix I thought you didn’t like charts without the zero bound on the bottom.

2nd, and far more importantly, if labor is having a bad time (which no one disputes) then “safe capital” is having a much worse one.

Show me a graph of how many hours of low wage labor the wealthy can buy with the interest on their 10-year treasury bills.

A simple interest on 1 million dollars in 1985 (113,800) was worth at least 6 low wage workers annual earnings.

By 1995 interest on 1MM fell to 77,800 call it 4 low wage workers.

By 2005 interest on 1MM was worth just $42,200… call it 2.5 low wage worker equivalents.

Today a million dollars will earn you roughly the same amount as a single laborer at the federal minimum wage.

Western labor (me) is having a tough go of it. Western capital (me too) has fared much much worse.

Posted by y2kurtus | Report as abusive

Hmmm! Even the Cleveland Fed can not agree on the labor vs capital scenario. Here is a paper by Paul Gomme and Peter Rupert, analysts for the Cleveland Fed. https://www.clevelandfed.org/Research/Po licyDis/No7Nov04.pdf

Their conclusions:
First, the “historic lows” in labor’s share are only observed in some series produced by the Bureau of Labor
Statistics. Other measures of labor’s share—for example, for the nonfinancial corporate business sector,
or the macroeconomy more broadly—are currently near their historic average.

Second, for labor’s share as computed by the Bureau of Labor Statistics, a fall in labor’s share does not
necessarily imply a rise in capital’s share; indirect taxes less subsidies constitute a wedge between these two
series. Consequently, a fall in labor’s share could be associated with a rise in capital’s share, but it could also be due to a rise in the share of indirect taxes less subsidies. However, as shown for the nonfinancial corporate business sector, the share of indirect taxes less subsidies does not vary much. Further, the terms
“capital’s share” and “profit share” are often used interchangeably, ignoring the fact that capital income
derives from more sources than just (corporate) profits.

Finally, there is a cyclical pattern to labor’s share: It rises during recessions and falls during expansions.
The recent fall in labor’s share—back to its historic mean—is typical of the early part of a business cycle
expansion. Whether these movements in labor’s share have implications for monetary policy is an open

Posted by DaveDopp | Report as abusive

I sent the question of “Does capital income include 401ks from the Baby Boomers?” and the response I got back confirmed it:


The initial contribution of a worker to a retirement account is part of labor income. In the following years, the investment return realized on the retirement account is part of capital income. (So capital income does include the realized returns on on baby boomers’ retirement accounts …)

Let me know if you have further questions. Best,


Filippo Occhino
Research Department
Federal Reserve Bank of Cleveland

Phone: (216) 579-2969
Fax: (216) 579-3050
Email: filippo.occhino@clev.frb.org


So the chart would be even more interesting if we had a chart of taxable income before we could come to the conclusion “There’s a big lesson here for anybody serious about fiscal policy, too. (Paul Ryan, I’m looking at you.) As the labor share of income goes down and the capital share of income goes up, the only way that we can stop tax revenues from plunging disastrously is to tax capital income at least as much as we tax labor income.”

Posted by EricPeters | Report as abusive

More @mlnberger than Felix, note that 1) individuals have continued to see their wages increase over the period of their working lives; each cohort is seeing lower wages with the requisite delay, and 2) since the guy before you mentioned demographics, it’s interesting to note that the incomes of different racial and ethnic groups have gone up faster than the overall level of income as the lower-income race/ethnicity groups have increased their share of the population and whites in particular have decreased their share of the population.

Posted by dWj | Report as abusive

Post Your Comment

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 http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/