Why I don't believe today's productivity stats

Posted by: Michael Mandel on May 07

Today the BLS put out a report saying that nonfarm productivity rose by 1.8% over the past year, and at an 0.8% rate in the first quarter of 2009.

To be blunt, I don’t believe these numbers. There are two reasons, both of which I have written about in the past.

*First, the productivity statistics don’t pick up drops in intangible investments such as R&D, advertising and marketing. For example, if companies are cutting back on R&D and firing all their scientists and technologists, that will show up as a drop in work-hours but not as a drop in output. Hence, it will look like an increase in productivity—but it’s really just us eating our seed corn.

Or to use an example that hits closer to home: Advertising. Advertising is not measured as part of output (it does not show up directly in either personal consumption or business investment). Therefore a plunge in advertising, like we are experiencing today, does not show up as a decline in output—but it does lead to a lot of my journalist friends being laid off. Poof! An apparent increase in productivity.

*Second—as I have written before, the productivity statistics are being distorted by mismeasurements of real imports. In fact, in areas like furniture, electronics, and autos, the government statisticians are overestimating the drop in imports, adjusted for inflation.

Why does this matter? Because you may not realize, for most of the economy the government has no direct way of measuring production. Instead, the statisticians measure how much people and companies are buying, and then subtract out imports. For example, retail stores report furniture purchases, and importers report imports of furniture from China and elsewhere. Then by subtracting imports from consumption, the difference is production.

The problem is that all of this has to be adjusted for inflation, and there are some real problems with the import price statistics. I am going to follow up with an extended explanation—too wonky here—but it tends to make the growth of production look higher than it is.


TrackBack URL for this entry: http://blogs.businessweek.com/mt/mt-tb.cgi/

Reader Comments

pushmedia1

May 7, 2009 11:45 AM

Productivity is defined as output per hour, so your complaints about hours dropping miss the point. If we produce more stuff with fewer labor hours, that's by definition an increase in productivity. It may not be the "good" kind of increase, but its an increase of that ratio.

You say that advertising doesn't affect output... then why do firms do it? Same with R&D. These aren't being calculated in output directly, but they're inputs to the production of final output (which is contracting).

Your import pricing problems are unlikely to be statistically important. Goods imports are 10% of GDP so the price changes would have to be off by about an order of magnitude for it to make a difference to the productivity numbers.

prb

May 7, 2009 11:46 AM

Yes, interesting that the 'stress tests' were delayed for some unknown reason. Then this week all reports 'point up". That is, less severe slowing.

prb

May 7, 2009 11:46 AM

Yes, interesting that the 'stress tests' were delayed for some unknown reason. Then this week all reports 'point up". That is, less severe slowing.

Mike Mandel

May 7, 2009 12:54 PM

Pushmedia

No, you miss the point. R&D and advertising are investments, like a new piece of equipment. They have long-lived consequences. So just like GDP counts investment in equipment and buildings as separate from today's consumption, it should also count R&D and advertising as a separate component to output.

The BEA is well-acquainted with this. In fact, it is gradually introducing R&D into the GDP statistics (I'll give you the link if you want).

dw

May 7, 2009 01:23 PM

how do we know that imports are only 10% of GDP? and why would they be any part of it to begin with? isn't the definition of an import, some thing that was produced in another country? which should part of that countries GDP not ours. after GDP stands for Gross DOMESTIC product doesn't it. or has it been so corrupted to the point it no longer measures US production?

pushmedia1

May 7, 2009 07:30 PM

Mike,

The satellite account is great, but its not nearly big enough to make qualitative differences in the productivity number. Services flowing from "stocks" of R&D will increase measured GDP by about 2%. Even if that decreased by half in this recession (an unlikely prospect), we'd still see increasing labor productivity.

Mike Mandel

May 7, 2009 09:46 PM

R&D is only one intangible. Please take a look at my story here.

http://www.businessweek.com/magazine/content/06_07/b3971001.htm

The other issue is the miscounting of imports.

pushmedia1

May 7, 2009 09:58 PM

Dw,

Imports aren't a component of GDP, they're subtracted off total expenditures (as Mike mentioned in the post). If they're undervalued, it will look like we have more GDP than we actually do.

My point is that this effect may well exist, its just not likely to be very big.

alec resnick

May 7, 2009 10:49 PM

Pushmedia1, I think the definition of productivity is less important than why we're looking to quantify it, in which case looking at the instantaneous output/hour ratio is the wrong way to go.

Mike, I'd be interested in the GDP link, or is the "GDP: What's Counted, What's Not?" article equivalent?

cm

May 8, 2009 02:31 AM

Mike: The fact that you don't believe today's productivity stats immediately begs the question whether you have believed any productivity stats prior to the pronouncement, or mounting evidence, of our current crisis, and if so then why? What difference in the methodology or environment would justify the different stance? (Other than the "productivity improvements" hitting closer to home now, but I don't want to be nasty.)

This from an engineer's perspective in "mature" corporate tech. Tech's heyday was largely before I entered the industry, but while "we" were enjoying our good time, it was manufacturing workers' turn to have their "productivity" "improved" by mass layoffs and offshoring, of late it's been ours, and now it shifts to your colleagues. I don't see how the worm will turn anytime soon.

Mike Mandel

May 8, 2009 08:27 AM

I have severe doubts about the earlier productivity stats too. Some of which I talked about in earlier articles (for example, "The Real Cost of Offshoring"
http://www.businessweek.com/magazine/content/07_25/b4039001.htm, and some of which I will be writing about in upcoming stories.

cm

May 9, 2009 01:49 AM

Thanks. No offense from my unfiltered remarks hopefully.

My largest experience and insight into business is in office-type environments. Given what I have seen and heard in and of US cubicle farms in about a decade, all claims of productivity growth have struck me as completely ludicrous. In large part, what office automation has accomplished is increasing the pace (and "feasible" volume!) of now digital paper-pushing. On a more serious note, computer improvements have enabled larger and more complex software (and IT-aided) solutions that were previously impossible due to limited storage capacity and processing speed.

Some of that has translated into better and/or more consistent quality in manufacturing and business process, and new mass-available products that we previously didn't have, another part means that we are now solving the same old problems in larger and more complex ways, or, as more of the human element has been cut out, in more rigid ways with more "process" enforcement and less wiggle room for judgement. Overall, one has to seriously question how this has improved the general human condition. But then this is no longer a productivity issue, but one of social policy.

Thank you for your interest. This blog is no longer active.

 

About

Michael Mandel, BW's award-winning chief economist, provides his unique perspective on the hot economic issues of the day. From globalization to the future of work to the ups and downs of the financial markets, Mandel-named 2006 economic journalist of the year by the World Leadership Forum-offers cutting edge analysis and commentary.

BW Mall - Sponsored Links

Buy a link now!