What's trade got to do with it?

Here is a fascinating graph from a new paper by my colleague Robert Lawrence. It compares trends in average productivity and real hourly wages between 1982 and 2006 in the U.S.

To say that wages have not kept up with productivity is an understatement.

Why this gap? Lawrence shows that some of it is a statistical mirage: actual labor compensation (which includes benefits) has risen more rapidly than wages; and the CPI which deflates the wage series has similarly risen more rapidly than the prices of the goods that workers produce (which is an interesting bit of empirical evidence relevant to an earlier debate in this blog). These two adjustments explain 60 percent of the gap. The rest has to do with rising inequality.

Lawrence also identifies an interesting change in the nature of the inequality over time. In the 1980s and to some extent the 1990s the driving force behind inequality was the inequality in the structure of wages, and the rise in the skill premium in particular. But more recently the skill premium has stopped increasing. The story since 2000 is essentially a shift in income from labor to capital.

Could globalization be behind these trends? Lawrence doubts it. While he is willing to attribute some of the rising wage inequality of the 1980s to trade with developing countries, the timing of the increase in profits since 2000 does not seem to fit very neatly with any globalization-related story. He prefers a cyclical interpretation, suggesting that the rise in profit shares will be reversed.

UPDATE: The link to the paper has been corrected. Thanks to Eric de Souza.

Comments

[reason]

Dani,
I saw an interesting argument (sorry cannot remember the source) that increases in household sector debt inevitably increase profit, since household expenditure is income for the corporate sector, and household income is an expense. Any comment - it could be relevant here, since presumably the credit cycle will reverse at some stage.

[mohammadreza]

sorry i have one question:what is inequality in the structure of wages?

[a student]

Does the share of profits have to go down? Perhaps now capital is more important to production.
This was also something that struck me about your "Democracies Pay Higher Wages" paper. Could it be that democracies, for whatever reason, encourage industries where labor is more important and so has a higher share of returns to output?

[Ktwdawg]

Domestic policy and regulation can affect wage inequality fairly directly, yes? If only I could think of some major political/idealogical shift in domestic corporate policy that would fit this timeline....

[a student]

Also, what are the units? I can't read them and I can't find the paper to check. Are we talking about a big increase in productivity here?

[-Lurker]

Can you provide the cite to the Lawrence paper that this graph appears in?

[Ivo Staub]

'The story since 2000 is essentially a shift in income from labor to capital.'

And from blue-collar workers to executives.

[Joe]

In yesterday's Independent, Stephen King (the best business commentator in the world!) writes:
"Those who own capital have good reason to be happy. Those in the West who can only offer labour - particularly those with relatively low skills - will sometimes feel only despondency." But he warns that a protectionist response from the West would be a tragedy.
http://news.independent.co.uk/business/comment/article2643120.ece

[save_the_rustbelt]

"Average" wages tell us nothing about the distribution.

I have no doubt average wages are up, I have no doubt the distribution is skewing to the top 20%.

[Miracle Max]

My colleagues have been documenting this for years (http://www.stateofworkingamerica.org/).

Nice to know it's official now.

[Eric de Souza]

The link to the paper is wrong. The correct link is:
http://ksghome.harvard.edu/~RLawrence/Lawrencefor%20Brandeis.pdf

[Eric de Souza]

Sorry, I should have been more precise. The link to the paper on Robert Lawrence's publications page is wrong. Your link sends one to his publications page. Clicking on the link there gives an error message.

[brad setser]

Dr. Rodrik -- Any chance you could spell out why Dr. Lawrence thinks the timing of the post 2000 increase in profits doesn't fit well with a globalization-related story. It certainly seems to match the strong acceleration in Chinese export growth which took Chinese exports from $200-250b a year to close to $1 trillion last yuear and likely over $1.2 trillion this year. But that is a very rough correlation -- I would be very interested in hearing a bit more of the thinking behind Dr. Lawrence's argument (about the timing not fitting well)

[Robert Lawrence]

Brad Setzer is correct. The timing of the increased profit share fits with the emergence of China and it could be that there has been a permanent increase. Indeed one reason corporate profits are so high is that real interest rates are low and this could reflect Chinese savings behavior. But there are also considerations that suggest that the low labor share in income is cyclical. First, the current share of labor is the same as it was in 1997 at a similar stage of the expansion. Second I would have expected the China effects to be evident particularly in manufacturing -- but the change in labor share is the same as in services. Third, why did earlier globalization not have the effect on labor share? Is China really all that different? My position is therefore that we will need to wait till the expansion runs its course before we can sure.

[Peter Schaeffer]

Could unabated mass immigration combined with an economic slowdown, be related to rising income inequality since 2000? I know this is heresy for the Open Borders crowd. But take a look at the following from Andrew Sum of Northeastern University

The Impact of New Immigrants on Young Native-Born Workers, 2000-2005
http://www.cis.org/articles/2006/back806.html

Over the 2000-2005 period, immigration levels remained very high and roughly half of new immigrant workers were illegal. This report finds that the arrival of new immigrants (legal and illegal) in a state results in a decline in employment among young native-born workers in that state. Our findings indicate that young native-born workers are being displaced in the labor market by the arrival of new immigrants.

Between 2000 and 2005, 4.1 million immigrant workers arrived from abroad, accounting for 86 percent of the net increase in the total number of employed persons (16 and older), the highest share ever recorded in the United States.

Of the 4.1 million new immigrant workers, between 1.4 and 2.7 million are estimated to be illegal immigrants. This means that illegal immigrants accounted for up to 56 percent of the net increase in civilian employment in the United States over the past five years.

Between 2000 and 2005, the number of young (16 to 34) native-born men who were employed declined by 1.7 million; at the same time, the number of new male immigrant workers increased by 1.9 million.

Multivariate statistical analyses show that the probability of teens and young adults (20-24) being employed was negatively affected by the number of new immigrant workers (legal and illegal) in their state.

The negative impacts tended to be larger for younger workers, for in-school youth compared to out-of-school youth, and for native-born black and Hispanic males compared to their white counterparts.

It appears that employers are substituting new immigrant workers for young native-born workers. The estimated sizes of these displacement effects were frequently quite large.

The increased hiring of new immigrant workers also has been accompanied by important changes in the structure of labor markets and employer-employee relationships. Fewer new workers, especially private-sector wage and salary workers, are ending up on the formal payrolls of employers, where they would be covered by unemployment insurance, health insurance, and worker protections.