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.

« Inflation Expectations, Almost Live, From The Cleveland Fed | Main | Selective Memory »

September 07, 2006

Another Bad Inflation Signal?

That would be the conventional interpretation based on yesterday's revision of growth in second quarter unit labor costs.  From MarketWatch:

U.S. workers have been more productive, but have also been paid more than previously believed, figures on Wednesday showed.

Revisions to quarterly nonfarm business productivity data show unit labor costs rose 5% in the past year, matching the fastest pace since 1990, the Labor Department reported. Read the full government report.
Higher unit labor costs could fuel inflationary pressures as firms struggle to recover their labor costs, and that could push the Federal Reserve to resume raising interest rates.
Unit labor costs -- the cost of the labor needed to produce one unit of output -- had been subdued in the past few years, but now workers are capturing more of their share of the productivity bonanza.
"We now unquestionably have an issue with wage growth," said Stephen Stanley, chief economist for RBS Greenwich. Higher wages mean consumer spending should hold up, but it raises troubling issues about whether a new inflationary spiral is beginning.
The figures "will keep the Fed in an uncomfortable position regarding the inflation outlook," said Michael Englund, chief economist for Action Economics.
A similar sentiment could be found at Reuters...

U.S. Treasury debt prices fell on Wednesday after a bigger-than-expected upward revision to second-quarter labor costs ignited fears that the Federal Reserve may need to resume its monetary policy tightening campaign...

"The major issue dividing the Fed is the danger posed by the bulge in labor costs, with some concerned that these costs will be pushed through into prices and others more confident that the monetary restraint the Fed has already put in place will prevent that from happening," said Pierre Ellis, senior economist at Decision Economics in New York.

"The numbers today really highlight that debate."

... and at The Wall Street Journal...

The latest data are "certainly good news for workers," said Stephen Stanley, chief economist at RBS Greenwich Capital. "They've been working against the strong head winds of high energy prices and a softening housing market. ... But these big revisions are something to be concerned about. I think we're in for a pretty extended period" of elevated inflation. (The data contributed to a drop in major stock indexes.

...and you get the idea.  But not everyone was so convinced that the rise in measured labor compensation -- specifically hourly compensation in the nonfarm business sector -- was all that meaningful.  Mark Thoma highlights comments by Paul Krugman, who attempts to repudiate any claims that labor compensation is rising by perpetuating a wrong-headed focus on narrow wage and salary data.

OK -- we'll ignore that one.  But Mark, along with knzn, also commends his readers to some quite sensible observations made by Dean Baker in a post last week.  Explains knzn:

The best explanation I’ve heard comes from Dean Baker who suggests, based on the NIPA statistical discrepancy, that some capital gains (obtained, for example, via exercise of employee stock options) might be misclassified as compensation. (Conceptually, in the case of stock options, the compensation took place when the options were granted, not when they were exercised. Anything that happened to the value in the intervening time was a capital change rather than income, but the value of the options doesn’t show up on the income side of the national accounts until they are exercised.)

I'm not sure about the misclassified part.  In terms of the national income side of things, any capital gains on the options may be a wash: We essentially have a transfer from firms (that would otherwise be able to sell their stock at market prices) to workers (who get the assets at a discount).  In terms of labor costs, those transfers certainly count.

But the general point is well taken -- a jump in capital gains might well have a relatively transitory impact on labor costs, and not be indicative of a trend.  Compare (as Dr. Baker suggests in his original post) the behavior of growth in the Employment Cost Index (ECI) with that in the hourly compensation series:




There are some differences in coverage between the ECI and nonfarm business sector compensation measures -- in the treatment of government and not-for-profit workers, for example -- but they are both broad measures that cover benefit payments as well as explicit wages and salaries.  But the hourly compensation statistic - and hence unit labor cost calculations -- includes stock options. The ECI does not.   

Caroline Baum writes today...

There was certainly nothing in the narrow measure of wages in the monthly employment report to suggest a 13.3 percent annualized jump in wages and salaries in the first quarter from the fourth quarter of 2005.

When Commerce's Bureau of Economic Analysis reported the revised figures last week, the lumping of the income gains in the first quarter suggested to many economists that the source was bonuses and options.

and Asha Bangalore seconds:

The most noteworthy part of the report was the significant upward revision of hourly compensation to a 13.7% gain in the first quarter from the previous estimates of a 6.9% increase. This is a reflection of compensation increases coming from stock options and bonuses which are reflected in the income side estimates of GDP. This is a one-off event and not a reflection of a big change in the underlying trend of labor costs.

That, and the stable-to-declining pattern of compensation in the ECI, should give us pause about concluding we are at the beginning of any lasting acceleration in the return to labor. 

September 7, 2006 in Inflation , Labor Markets | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Another Bad Inflation Signal? :

» Is Labor Compensation on the Rise? from Economist's View
David Altig looks into whether recent upward revisions in unit labor costs represent another good wage signal or, as he prefers, another bad inflation signal. Okay -- ignore that one, but David does say sensible things, for the most part [Read More]

Tracked on Sep 8, 2006 4:50:19 AM


I’m going to argue the point about misclassification. Suppose that, instead of giving call options, the company gave actual stock, along with put options, and some reduction in cash compensation to account for what would have been the strike price of the call option. If I remember my options theory correctly, this is exactly equivalent. But when the stock price subsequently goes up (before the call option would have been exercised), the increase in the stock price would not be classified as compensation. (Nor should it be, it seems to me, since the company could have paid the initial value in cash, and the employee could have chosen to purchase the stock just like anyone else.)

Or suppose that, instead of granting an option to the employee, the company wrote a call option, sold it at the market price, and paid the proceeds to the employee. Then suppose the employee purchased the option on the market. Again, this is exactly equivalent (except for the restrictions on the option), but the company’s subsequent losses on the option would not be classified as compensation. (After all, why does it matter if the employee bought the option or somebody else bought it?)

I would agree that the initial value of the option (which includes any initial in-the-money-ness at the time the option is granted) is correctly classified as compensation, but that’s not what I understand Dean Baker to be talking about.

Posted by: knzn | September 08, 2006 at 02:36 AM

knzn -- Good points. The definition of capital gains is the sticky issue here. When I go back and read Dean's piece, he relates capital gains income to the exercise of stock options. Since we both agree that the grant of options ought to be viewed as compensation, it seemed to me that his comments were more about mis-timing than misclassification. In other words, Dean's (and your) connection to of the exercise of options to the statistical discrepancy in the National Income and Product Accounts would exist even if there were no true capital gains.

So how to think about a true capital gain? That, I agree is trickier. I like the second example you have give: If the company writes a call option and sells it on the market, paid the proceeds to the employee, and the employee purchases the option on the market, subsequent losses on the option are not treated as compensation. A comparable example might be insurance -- suppose that a firm self-finances its health care benefits. There will be some acturially fair value of that benefit, but in any given year the actual benefit to workers may well be higher than that value. (And sometimes lower -- there will be times when the option is never exercised!) Should we view the value of the compensation in terms of the ex ante benefit or the ex post benefit?

As I think it about in terms of economic decision making, I am persuaded that the ex ante benefit is right: Firms will hire workers up to the point where marginal costs and marginal returns to labor are equated in expectation. When I think about the accounting aspect -- simply adding up how much of national production accrues to labor and how much to capital -- I'm less certain.

Great comment.

Posted by: Dave Altig | September 08, 2006 at 08:14 AM

Sure, options are in fact compensation for labor performed. But they are incentive compensation, and they go mostly to high-ranked supervisory employees and executives. While they are indeed "earnings," to count option gains as "hourly earnings" is misleading in the extreme. To add an exclamation point: salaried workers are the ones who receive stock options, and "salaried" workers are NOT hourly workers!

In the matter of statistical discrepancies: it is really hard to imagine that corporate profits as reported quarterly by individual companies, could be rising as fast as they are if hourly wages are skyrocketing. Something doesn't compute here.

Posted by: JB | September 09, 2006 at 12:10 PM

The effect of options based compensation on the issuing corporation is actually very simple. The company is giving to the employee, a percentage ownership share in the business. Measurement of that share may be a bit complicated --you are issing options on shares of stock rather than actual shares. Thus one needs Black-Scholes or some other mathematical formula to calculate the expected number of shares that the option grant represents. In the end, though, I repeat: in an options grant the company is giving potential shares of stock (which represent a percentage ownership share of the business) to employees rather than giving them a cash bonus.

Do options grants cost the other shareholders money? You bet they do. You would certainly not give away a 1% ownership share in your house (or even a potential 1% share that had a decent probability of becoming real) without compensation, would you? Dilution of shareholder equity sounds abstract but it is a very real transfer of a percentage of ownership in the corporation, from outside shareholders to employee shareholders. That is why many investors regard options based compensation as theft.

Do options grants contribute to cost push inflation? Fuggedaboutit!! Do you really think cost accountants figure the cost of options grants into product costs when companies set prices for their goods and services? If you do then you have no understanding of how Corporate America works. Suffice it to say this: the whole idea of options based compensation is to give money to employees without drawing on corporate cash and while also minimizing the effect on reported earnings.

Posted by: JB | September 09, 2006 at 01:09 PM

JB -- Thanks for the comments. It does turn out to be the case that the connection between unit labor costs and price-inflation is not as tight as theory suggests. I think Caroline Baum made this point in her column.

Posted by: Dave Altig | September 10, 2006 at 08:18 AM

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