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 Yuan Float, One Year Later | Main | Odds And Ends -- July 22 Edition »

July 21, 2006

What Does Labor Compensation Measure?

In one of my posts yesterday I staked out an old position of mine: Payments made to workers should be broadly measured, including both explicit monetary remuneration and non-cash benefits.  Not surprisingly, I received several comments arguing that increases in labor compensation have, in the recent past, been driven by rising health care costs.

In making sense of all this, I think it helps to consider two different questions we might want to ask when considering payments to workers.  The first is, "What does it cost a business to purchase one hour of a worker's time?"  To answer this question, a broad (benefit inclusive) measure of labor compensation is clearly the right one.

Now consider the second question: "Does an increase in payments made by firms to workers make workers better off?"  On this issue, the answer may very well depend on the difference between compensation paid in dollars and compensation paid in benefits.

Maybe, but not necessarily.  Consider two workers, both of whom receive the same total amount of compensation, and both of whom spend exactly the same amount on health care.  Now suppose that those expenditures rise by 10% for both workers.  We'll further suppose that worker 1's wages rise by 10%, while worker 2's employer holds his or her salary constant but picks up the tab for the higher medical costs.  Would you say that worker 1 is better off than worker 2?  Neither would I.

The issue, of course, is that, left to their own devices, worker's would not choose to purchase the same quantity of health care when prices rise.  They therefore would not be indifferent between a dollar received in cash -- which they could allocate as they wish -- and a dollar received in health care benefits -- which they have no option to spend any other way.

That is an argument to which I am extremely sympathetic, and I am all for policies that disentangle the provision of insurance from employment. (A good debate on these issues can be found here.)  But it is an issue that has little to do with macroeconomic performance.

So, if you what you are interested in is what sorts of payment arrangements generate the highest level of well-being for employees, using narrow wage-and-salary measures of labor compensation may be justified (though not totally -- those benefit payments probably still yield utility).  But if what you are talking about is how well the economy is doing in generating income for workers, an inclusive measure of labor compensation is a must.

UPDATE: Although you probably read it there before reading it here, Greg Mankiw shares my sentiment, Brad DeLong does not. Brad does make the point -- I'm paraphrasing -- that if you are worried about the distribution of labor compensation, and if the payments to people in the part of the income-distribution you care about are well-captured by average hourly earnings, that would be a reason to prefer the series that excludes fringe benefits.  Fair enough, but I'll repeat my earlier argument that it would serve everyone well to be explicit that we are than using a notion of economic well-being that discounts average performance and focuses on who gets what. 

Meanwhile at Angry Bear, I believe that, as is often (though not always) the case, pgl and I have converged.   

UPDATE, AGAIN: The Liberal Order suggests:

... let's not also forget to account for workplace safety regulations and other observed changes in safety. For example, the manufacturing jobs that exist today are on average much safer than those of thirty and forty years ago. The increases in safety act as a compensating differential. This means that increases in productivity must also be weighed against increases in workplace safety.

A key issue is whether these changes add to the marginal cost of labor or simply the average cost of labor (as would be the case, for example, if safety provisions are more like fixed capital).  It's an interesting point, and an interesting question.

UPDATE ONCE MORE: knzn has some thoughts as well.

July 21, 2006 in Labor Markets | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference What Does Labor Compensation Measure? :


What you need to bring into the analysis to understand the comparisons is the tax impact.

In your example the person who pays for his own insurance has to pay for it out of after tax dollars while the individual that
has his insurance paid for by his employeer is receiving the benefit on a pre-tax basis.

In your example where one individual gets a wage increase to offset the 10% rise in insurance costs will be worse off because he will have to pay income come tax on the income and so will not get an aftertax increase large enough to offset the rise in health insurance costs.

The reason insurance is entangled with employment is because the tax treatment make it cheaper for the employeer to provide insurance as a benefit. If the tax rate is 25% the employeer can buy $75 worth of insurance for $75 on a pre tax basis but must pay the worker $100 for the worker to be able to buy $75 worth of insurance out of his aftertax income..

Posted by: spencer | July 21, 2006 at 04:07 PM

What about pension and retirement benefits that are promised but not delivered? These aren't even accounted for on the paycheck stub, so cutting back doesn't show as a loss.

Posted by: Lord | July 21, 2006 at 04:34 PM

Dave - "But if what you are talking about is how well the economy is doing in generating income for workers, an inclusive measure of labor compensation is a must."

Ok. Let's do it for private industry.

Results: First quarter 2006 real compensation per hour is in negative territory, along with real wages, for private industry employees.

Employment Cost Index news release
March 2005 Release

"Compensation costs in private industry rose 2.6 percent in the year ended March 2006, slowing from a 3.5 percent increase in March 2005."

But real compensation is down compared to the 2005 annual rates for most if not all employment sectors. (And that's without challenging the manner in which some compensation benefits are calculated, including healthcare insurance.) Services may be a slight exception. But very slight.

"Private industry workers wages and salaries increased 0.7 percent during the March 2006 quarter, compared with a 0.6 percent gain in the previous quarter. Private sector benefit costs rose 0.4 percent for the March quarter, following a 0.7 percent gain in the previous quarter."

But real wages are down. And Bernanke, unfortunately, was off the mark with his testimony before Congress yesterday on this point. Real wages are down. Period.

The 12-month percent changes in wages and salaries for each of the major employment sectors are in negative territory for the second year in a row, ranging from -0.4% to -1.0%.

Employee compensation benefits continues to show growing weakness.

Benefit Costs, Private Industry, 3-month percent changes:

Jun 2004 - 1.6%
Sep 2004 - 1.0%
Dec 2004 - 1.2%
Mar 2005 - 1.5%
Jun 2005 - 0.8%
Sep 2005 - 0.9%
Dec 2005 - 0.7%
Mar 2006 - 0.4%

National Compensation Survey - Compensation Cost Trends

Select this interactive link:
Create Customized Tables (multiple screens), NAICS basis

2006, Otr 1 vs. 2005, Qtr 4
12-month percent change
Value of Compensation: Total benefits, not seasonally adjusted:

All workers - 3.0%, down from 4.0% in Qtr 4, 2005
All union workers - 2.9%, down from 3.3%
All nonunion workers - 4.2%, down from 6.0%
Blue-collar occupations - 2.1%, down from 2.9%
Management, professional, and related - 3.2%, down from 4.8%
Sales and office - 3.3%, down from 4.4%
Service occupations - 3.3%, up from 3.1%
Natural resources, construction, and maintenance - 3.2%, down from 3.7%
Production, transportation, and material moving - 1.4%, down from 2.4%
White-collar occupations - 3.2%, down from 4.6%
Manufacturing - 0.7%, down from 4.2%
Aircraft manufacturing - (-18.5%), down from 40.6%
Goods-producing industries - 1.3%, down from 3.8%
Service-providing industries - 3.5%, down from 4.1%

First Quarter 2006, revised
1 June 2006 release

Review Table 1.

Real Compensation Per Hour:

Business sector - 121.5, down from 2005 annual rate of 121.6
Nonfarm business sector - 120.6, down from 2005 annual rate of 120.8
Manufacturing sector - 126.8, down from 2005 annual rate of 128.2
Durable manufacturing sector - 123.9, down from 2005 annual rate of 125.1
Nondurable manufacturing sector - 130.6, down from 2005 annual rate of 132.4
Nonfinancial corporate sector - 118.5, down from 2005 annual rate of 118.7

Posted by: Movie Guy | July 21, 2006 at 05:30 PM

We have converged.

Posted by: pgl | July 22, 2006 at 10:06 AM

Counting safety regulations as worker compensation is ridiculous. These are to reduce the employers expense of lawsuits from injuries. The employee may benefit from not being injured but it is the employer that benefits from not have to pay for it.

Posted by: Lord | July 22, 2006 at 01:16 PM

«Counting safety regulations as worker compensation is ridiculous.»

Counting safety regulations is a fantastic way of doing a ''hedonic'' adjustment of wages, like for the various price indices. Very politically useful to argue that even if salaries are doing not that well and prices are doing pretty well, the cost of staying alive is not becoming worse.

By the same token, any extension of life expectancy should be taken as extra compensation too...

And of course what about the present monetary value of having an administration with such a record of tough anti terrorist wars? That is worth a lot of money to all those feeling better protected. :-)

«These are to reduce the employers expense of lawsuits from injuries.»

Well, the theory of course is that a safe workplace is not a right, but a valuable concession that generous businesses voluntarily bestow on workers. After all, workers are instantly replaceable, and businesses could instead just fire any partially or terminally damaged ''bulk headcount''... :-)

Or better, businesses could encourage accidents to collect faster on their insurance policies:


But with civic generosity they grant their workers raises in the form of reduced workplace danger.

I think out century badly needs a MarkTwain or as someone said a Mencken to address the amazing logic from some quarters...

Posted by: Blissex | July 23, 2006 at 05:00 PM

Spencer -- My example was just that: An example. You are, of course identifying why we attache health benefits to employment, a practice that I think is not a good idea.

Lord and Blissex -- Any payments made as a result of employing one more unit of labor represents a relevant cost of labor, independent of whether the government taxes and throws it away, whether the worker values payments made on their behalf the same as cash, or whether we think those costs ought to be incurred by firms for good social reasons. The problem with treating OSHA type regulations as compensation is not the economic principle, but the probability that such regulations do not represent a *marginal* cost to labor.

MG -- I have no beef with analysis like yours. If everyone would do as a good a job, we's have a good basis for further discussion.

Posted by: Dave Altig | July 24, 2006 at 07:17 AM

Any payments made as a result of employing one more unit of labor represents a relevant cost of labor

But you would also have to count the costs of lawsuits and offset reductions in the growth of such to the cost of safety regulations. Counting costs but ignoring benefits, one cannot identify true costs.

Posted by: Lord | July 24, 2006 at 01:54 PM

The value of pension and retirement benefits that have been jettisoned over the past, say, 25 years, probably amounts to 10-20% of income, so there probably hasn't been much true real growth of compensation over that time period. What we have been measuring as growth is largely borrowed future consumption.

Posted by: Lord | July 24, 2006 at 02:08 PM

More to the point, the "cost of safety" is like the "cost of quality". It is not safety or quality that cost, it is the lack thereof. Rather than speaking of the "cost of safety", we really should be speaking of the "benefit of safety". It is likely to have contributed to the productivity and profits we see.

Posted by: Lord | July 24, 2006 at 06:34 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