The Atlanta Fed's macroblog provides commentary and analysis on economic topics including monetary policy, macroeconomic developments, inflation, labor economics, and financial issues.

Authors for macroblog are Dave Altig, John Robertson, and other Atlanta Fed economists and researchers.

« The Employment Report And The Expected Funds Rate Path: Not Quite What I Expected | Main | The Trade Report »

January 10, 2006

What To Think Of The Labor Market

Here's what we know: Subject to some further revision, employment growth for 2005 will top two million jobs for the first time in five years.  What we don't seem to have consensus on is what to think about that number. Opinions have been mixed on whether this is a cause for celebration, or a source of continued concern about the strength of the US economy. Daniel Gross is firmly in the latter camp, a fact he reiterated in two posts yesterday.  In the first, he borrows a page from Angry Bear's pgl:

... the most recent jobs report from the Bureau of Labor Statistics? The employment-population ratio is 62.8 percent, lower than it was for all of the second half of the 1990s. This is not what full employment looks like.

In the second, he samples from a column by Stephen Roach:

Unfortunately, for the American worker, this jobless recovery has also been “wageless” -- characterized by an extraordinary stagnation in real wages.

I will discuss the employment-to-population ratio in a subsequent post, but let me try just one more time to drive a stake into the heart of the "wageless recovery" canard. 

As I argued in a previous post, the-wage-sky-is-falling claim is almost always based on the Bureau of Labor Statistics' real average hourly earnings series which is, frankly, the wrong thing to be looking at if we want to learn about the growth of labor compensation generally.  In that earlier post I emphasized the absence of nonwage compensation in the series, and showed that the fate of workers looks much different if we focus on a measure like real compensation in the nonfarm business sector, which does not exclude payments in the form of fringe benefits. Daniel is smart enough to anticipate this argument, and (in his first post) qualifies Roach's not-so-nuanced claim with this one:

Oh, and average compensation has been rising faster than inflation only because health care inflation has been off the charts for several years. So people who work get (1) lower cash wages; and (2) the same or somewhat less in terms of health benefits.

That statement has the advantage of not falling into the trap of ignoring the nonwage compensation element of payments to workers.  But it retains the disadvantage of being wrong. 

This time I will turn to labor payments as represented by the Employment Cost Index (ECI), which I do because it allows us to look at wages and salaries separately, and because my labor-data-expert colleague Mark Schweitzer likes it the best of all the alternative measures.  (One significant way the ECI differs from nonfarm business compensation, which is the series I used in my previous post, is that the ECI does not include compensation associated with stock options.) 

Here, as before, is the total compensation story, conveniently broken down by each presidential term since 1981:


No question -- the past year is not comforting.  But stacked up against its immediate predecessors, the first Bush term looks pretty good by this measure.

Gross's point, of course, is that wages and salaries alone paint the truer picture of how well workers have fared. OK, let's go to the videotape:


No falling wages here (the GHW Bush and Carter years excepted).  Those who say that wages have been falling steadily since 2000 are invariably basing their claim on real hourly earnings data that applies to production and nonsupervisory workers only (a point made by Jim Hamilton). The relative wages of that subset of workers have been trending lower for a long time, giving a distorted view of how labor is faring more generally.   That trend disappears when we focus on the wage and salary payments made to the larger set of workers captured by the ECI.

Although 2005 is nothing to brag about, the assertion that wage and salary payments fell broadly in the period from 2001-2004 is simply not supported by the facts.  And stacked up against other four-year periods since 1978, the record of the first Bush term is not particularly unusual, let alone unusually woeful.      

Commenting on my first post, Brad DeLong made the quite correct observation that labor compensation was weak during the period in question if measured relative to productivity growth. There you have a case.  The following table shows the difference between productivity growth and growth in total compensation for the nonfarm business sector.  A larger number, therefore, indicates that less of any realized productivity gains were enjoyed by workers:


It is a fact that the first Bush term takes the 30-year prize for the lowest labor compensation gains relative to productivity growth.  Whether it is substantially worse than, for example, the first terms of Reagan and Clinton -- well, I'll leave that to you. 

To my eye, there is plenty of ammunition here for both supporters and detractors, and there is an excellent ongoing discussion to be had regarding what, exactly, is going on with labor markets.  But the first step toward making that discussion productive is getting the facts straight.

January 10, 2006 in Labor Markets | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference What To Think Of The Labor Market :


I've come across some surprisingly good news in my report:

Grab a Hardhat (one with a headlight)

Posted by: Tim | January 10, 2006 at 06:05 PM

"Those who say that wages have been falling steadily since 2000 are invariably basing their claim on real hourly earnings data that applies to production and nonsupervisory workers only"

Can you help characterize the scale of the error? Who are the workers that are counted by ECI and not real hourly earnings? White collar managers? What fraction of the workforce do they represent? What fraction of total compensation isn't being counted?

Also, the WSJ mentions today that CEO compensation has been on a tear for the last decade or so. How much are the ECI numbers affected by increases in compensation for top managers?

Posted by: Anon | January 11, 2006 at 11:37 AM

o.k. o.k. I accept that the real wages have not been falling since 2000. But look at the trend in real wage growth (using ECI and PCE). Wages were growing faster during the recession than now. I guess it is more informative to look at these numbers quarterly than by presidential term.

Also, fringe benefits may be rising but how does that help me pay the mortgage?

2001 - Q1 1.55623
2001 - Q2 1.34027
2001 - Q3 1.63354
2001 - Q4 1.98067
2002 - Q1 2.36595
2002 - Q2 2.30162
2002 - Q3 1.64054
2002 - Q4 1.08088
2003 - Q1 0.6584
2003 - Q2 0.91534
2003 - Q3 1.10915
2003 - Q4 1.11772
2004 - Q1 0.57541
2004 - Q2 -0.23135
2004 - Q3 -0.18601
2004 - Q4 -0.66057
2005 - Q1 -0.27446
2005 - Q2 -0.16433
2005 - Q3 -0.80192

Posted by: Secret Admirer | January 11, 2006 at 11:55 AM

We have to face that the economy and employment suck most of the time, although we believe they can be much better.

Posted by: Lord | January 11, 2006 at 01:24 PM

Anon -- This is a good source for the sorts of questions you are asking: http://www.bls.gov/opub/mlr/2005/05/art1full.pdf

Posted by: Dave Altig | January 11, 2006 at 02:41 PM

We can debate the compensation v. wage metrics. We can also put forth different views as to the meaning of the lower population to employment ratio (note the page Gross got from me simply noted the data). But there is no debate as to the nonsense emanating from S. Mallaby's WaPo oped. Gross was right to blast its spin - even if one thinks Gross had his own spin in the rebuttal. I think Gross's point is simply this: certain WaPo reporters merely parrot whatever the White House tells them regardless of how untrue the message might be. For more on this - see what Kash posted today about the latest message from the Sec. of Dishonesty (eh, I meant Treasury) Jack Snow.

Posted by: pgl | January 11, 2006 at 03:05 PM

Macroblog or others:

what about trends in real wages in illinois? my rant and cited statistics are at the URL below.


Posted by: nate | January 12, 2006 at 12:11 PM

Who knew that "a job for everyone that wants one" was to be resolved by ensuring no one would.

Posted by: Lord | January 12, 2006 at 01:45 PM

pgl -- Well, as you know I stay out of the business of commenting on political spin. Bloggers are a different story -- when I goof up I expect people to call me on it, and any way you cut it, what Dan said was wrong.

nate -- I worry about the median income reports for reasons that are similar to my criticisms of hourly earnings. In particular, I think the median income numbers you make reference to include only money income. Because of this, I am suspicious that they are giving us the complete picture.

Lord -- my, so cynical.

Posted by: Dave Altig | January 12, 2006 at 04:48 PM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

Google Search

Recent Posts



Powered by TypePad