Let's Chat Labor Productivity

One might notice the comment, Woe to the U.S. worker when productivity metrics are reported. Over and over again, I note that offshore outsourcing is the taboo word among mainstream economists, regardless of the numbers.

Well, it seems I am not alone in that assessment. A New York Times op-ed drives home the point:

There’s a problem: labor productivity figures, which are calculated by the Labor Department, count only worker hours in America, even though American-owned factories and labs have been steadily transplanted overseas, and foreign workers have contributed significantly to the final products counted in productivity measures.

The result is an apparent drop in the number of worker hours required to produce goods — and thus increased productivity. But actually, the total number of worker hours does not necessarily change.

This oversight is no secret: as Labor Department officials acknowledged at a 2004 conference, their statistical methods deem any reduction in the work that goes into creating a specific unit of output, whatever the cause, to be a productivity gain.

This continuing mismeasurement leads economists and all those who rely on them to assume that recorded productivity gains always signify greater efficiency, rather than simple offshoring-generated cost cuts — leaving the rest of us scratching our heads over stagnating wages.

See that folks? Indeed, other Academics have quantified these errors and coined a new term, phantom GDP.

Economist Robert Gordon, Northwestern University, wrote a paper, Okun’s Law, Productivity Innovations, And Conundrums in Business Cycle Dating.

In it, he shows Okun is broke. He also offers up some modifications to fix it.

Firstly, What is Okun's law? It's kind of a rule of thumb, or ratio on how much one can expect the unemployment rate to decrease when the real GDP increases.

Or taking the description of Okun's Law from Gordon's paper:

A rough prediction that a cyclical deviation of output from trend would be divided between a 2/3 response of aggregate hours and a 1/3 response of productivity.

In turn, the Okun characterization subdivided the 2/3 hours response into 1/3 for the employment rate, 1/6 for hours per employee, and 1/6 for the LFPR (labor force participation rate).   

Now what is interesting on Gordon's paper is he sees a clear break, through regression analysis, of Okun's law around 1986. One of the events during this time was increased trade. 1986 also was the year global tariffs were lowered and the WTO was created.

I suggest you read the paper, attached below, but for now, Okun's law holds up to 1986 and after....breaks.

First I would like to show one of Gordon's conclusions:

The paper suggests a set of complementary hypotheses to explain this change in behavior. The overall shift in structural responses after 1986 is linked to the increase in the inequality of the income distribution. The declining minimum wage, the decline of unionization, the increase of imported goods, and the increased immigration of unskilled labor have undermined the bargaining power of American workers. As a result, employers can reduce labor hours with impunity and without restraint in response to a decrease in the output gap, in contrast to the period before 1986 when their behavior was more constrained by the countervailing power of labor.

Blog speak translation: U.S. workers are now considered dirty diapers, commodities, disposable and replaced easily by putting an ad on monster.com or through people importation.

Dr. Gordon also has other correlations, especially executive compensation incentives and stock options motivations, but the real issue is how our government, corporations and even economists simply will not, will not examine the O word, that being outsourcing or trade to explain jobless recoveries.

He also dismisses various theories to try to explain away soaring productivity, declining hours per worker and a negative job gain for this past decade, with some simple well reasoned facts.

He also analyzes data from a different filtering method. He uses a Kalman filter, which for all you geeks out there, a phased lock loop is in this category, instead of a Hodrick-Prescott filter. Without going into the properties, let's just say a Kalman is like a GPS unit to figure out where you are in a business cycle whereas a Hodrick-Prescott is using a map drawn in 1950. But here's the deal, around 2002, trend lines diverge for business cycles and they interpret these results as a structural change.

One thing they discovered, and you've seen this result many times in comparison graphs here, the 2008-09 slump is worse than 1980-83 for the labor market but not quite as bad for output (real GDP).

The paper then continues on analyzing the output gap, (which is a real fancy way of sayin' how much the economy could have been growing (potential GDP) if we didn't have the damn Banksters causing a recession, and what's really going on with economy (actual GDP)) to the unemployment gap. Again, ignoring the mathematics and modeling, we get this:

This indicates that changes in the cyclical deviation of productivity growth are highly noisy and negatively correlated, consistent with the view that short-term productivity fluctuations are a residual reflecting lags in hours behind output rather than as an exogenous technologically driven process as imagined in the RBC model literature.

Blog speak translation: Those guys trying to blame job losses on advances in technology exclusively are full of shit.

There is consistently a little productivity bubble or increase in productivity at the tail end of a recession, but nothin' like we've seen since 2002.

The employment rate declined more and the unemployment rate increased more in 2008-09 than would have been expected, given the output gap and given average responses over the 1986-2009 period.

Blog speak translation: Ignoring global labor arbitrage gets egg on the face of the White House with low ball unemployment projection numbers.

Finally the real bomb, buried deep in the paper:

The perverse role of the internet in causing an increase of labor hours responsiveness may be related to the Autor polarization hypothesis that middle-level white collar employees have been turned into mere commodities by the ubiquity of substitution between people and computers, both at home and overseas.

Blog speak translation: People are being treated as a commodity, easily replaced, by cheaper labor, facilitated by technology, locally as well as offshore outsourcing and is causing a Race to the Bottom on wages. (Here is the Autor polarization hypothesis (pdf).

Because of Dr. Gordon's meticulous mathematics and theory, his previous paper on productivity discounted offshore outsourcing and it's affects, but it is not really his fault per say. Instead, it is a good example of bad inputs. He used Goldman Sachs beyond belief low ball estimate of only 200,000 jobs offshore outsourced.

This is obviously false, the current estimates for the BPO industry alone (Business Process Outsourcing) this year is $150 billion dollars and expected to increase at least 10 fold in the next ten years.

Gordon does refer to technological innovations, which are super cheap communications and Internet bandwidth which did enable offshore outsourcing on steroids as well as corporate executives who would kill their own mother (in other words fire all of their employees) to keep a buck in their pocket, in 2003.

What does this tell us? Our government not only needs policies to stop offshore outsourcing and global labor arbitrage, it also states we need accurate data. They should be collecting on how many jobs are offshore outsourced, how many contracts, private, local, State and Federal are offshore outsourced, how many guest workers are in certain occupational categories in the U.S. and also collect on immigration status of all workers in the United States. Reality is we're in a global world and sticking your head in the sand to the point economic laws break is not getting us anywhere.

Previously we discussed divergence of GDP growth to the unemployment rate in Broke Okun.

I started this post this morning, having bookmarked Dr. Gordon's paper as a topic weeks ago. Amazingly enough Okun was covered by the San Francisco Federal Reserve today:

If productivity keeps on growing at an above-average pace, then unemployment forecasts based on Okun’s law could continue to be overly optimistic.

Trade Reform also gives a layman's description as does Angry Bear.

PDF icon Gordonprodokunbuscycle100108.pdf1.36 MB