The Problem With Productivity Growth

It can't accelerate forever, which means wage gains and labor tightness must eventually prevail.
Publish date:

Hilton Head? Thpft. Come to Butt-head.


Gross domestic product

(or GDP)

rose at a 2.3% annual rate during the second quarter. The table below shows GDP increases, alongside which sectors of the economy added to (a) or subtracted from (s) growth, for both the past two quarters and the past two years.

Consumption rose at a 4% annual rate during the second quarter, for example (this number is not shown in the table), and, because it accounts for roughly 69% of GDP, it contributed a little more than 2.7 percentage points to the second-quarter growth rate. Also, recall the familiar equation that GDP = C + I + G + X and, looking at the second-quarter numbers, note that 2.3 equals (roughly) 2.73 plus 0.52 minus 0.21 minus 0.75. Finally, note that investment comes in four forms: business investment (computers and peripherals and information processing and such), residential buildings, commercial buildings and inventories. Again, looking at the second-quarter numbers, note that 0.52 equals 1.18 plus 0.23 minus 0.03 minus 0.86.



There are two things to take away from the second-quarter GDP release.

(a) Final demand was still strong when the third quarter began.

The jackasses on television this morning -- the ones whose analysis clearly begins and ends with looking at the headline GDP number -- didn't pick this up, and that's (presumably) why suckers like y'all pony up for access to a stupid column on a site like this.

For original analysis and unique insight? For things of real value? For things that will genuinely aid decisionmaking?


Something called

final sales to domestic purchasers

(or FSDP) is arguably the most important line in any GDP report. FSDP equal personal consumption expenditure plus investment plus government spending (you can find this beast in Table 1 of the GDP release). In plain English, FSDP represent the broadest available measure of final domestic demand.

On a real (inflation-adjusted) basis FSDP grew an average 4.2% per year during the 1996-1998 growth boom.

They were growing at a 5% rate when the third quarter began.

On a nominal basis FSDP grew an average 5.5% per year during the 1996-1998 growth boom. They were growing at a 6.3% rate when the third quarter began.

Can you really interpret those numbers to mean that the economy is slowing significantly?

You can if you're long bonds. Or peddling shares.

(b) Price pressure is increasing.

The chain-type

price index for gross domestic purchases

(or PIGDPU) measures the prices of everything (including imports) Americans buy. It is the broadest available measure of the prices paid by U.S. residents (you can find this beast in Appendix Table A of the GDP release).

This inflation measure put in a bottom over a year ago, when it fell 0.2% during the first quarter of 1998. It then went on to rise 0.4% during the second quarter, 0.7% during the third and 0.9% during the fourth.

Then it rose 1.2% during the first quarter of this year.

Then it rose 2.1% during the second.

Also note that the core (excluding food and energy) PIGDPU shows a similar pattern of progressively bigger increases, as does the chain-type

price index for personal consumption expenditure

and its core counterpart (those numbers also appear in Appendix Table A).

In short, then, the GDP report shows that growth is barreling along and product-price pressure is increasing.

And so, too, is labor-cost pressure.


Employment Cost Index

(or ECI)

measures the change in the cost of labor. It includes both wages and salaries (roughly three-fourths of the index) and employer costs for employee benefits (the other fourth). The year-on-year increases revealed in the second-quarter ECI report released this morning are included in the table above. There are two things to take away from the numbers.

(a) The first-quarter deceleration in the benefit-costs portion of the index sure looks like a head fake.

When the first-quarter ECI report was released back in April, it was tough to believe that two-years-running acceleration in benefit costs had suddenly reversed course for no apparent reason -- especially because the

consumer price index for medical care

(or CPIMC, which tracks benefit costs pretty well) wasn't showing any deceleration.

The second-quarter (re)acceleration in the benefit-costs portion of the index puts it back on its upward track, one that better squares with the CPIMC and with what employers are reporting anecdotally about both health-care costs and benefit costs in general.

(b) Wage growth must be troubling the Fed.

Growth in even the sorrier measures of compensation -- neither the wages and salaries portion of the ECI nor the average hourly earnings series from the

employment report

includes stock options -- is approaching 4%, and growth in the better ones has been running above that pace for about a year now.

Indeed. The measures of hourly compensation that are released alongside the published


data (and which do include stock options) now show year-on-year increases as big as 4.4%. So, if the bigger productivity numbers we've seen over the past few years are to be believed -- and the Fed most certainly does believe them -- then so too should the bigger compensation numbers that go with them.

And the tighter the labor market gets, the more wages will accelerate.

And the more wages accelerate, the more productivity will have to accelerate to keep the Fed from tightening.

And goshdurnnit if that just doesn't seem very likely.

Side Dish

Why do you pay for this site?

Having trouble sleeping.

Keen focus on family values.

I guess I am just stupid that way.

Unparalleled micro-cap and mutual-fund coverage.

For fascinating personal details about the lives of the writers.