Sunday, October 03, 2010

Increases in U.S. Worker Productivity, More Than China's Currency, Responsible for Loss of U.S. Jobs

According to AFL-CIO President Richard Trumka:

"The Chinese government has been keeping the renminbi undervalued by about 40% as a deliberate and central plank of its export strategy. That's a 40% subsidy for the products they send here and a 40% tax on products we try to send there. Our dollars stimulate another nation's economy while we rack up unsustainable trade deficits at home.

The result is writ large on the walls of the Great Recession: 57,000 U.S. manufacturing facilities shut down in the last 10 years, costing us five-and-a-half million good-paying jobs, nearly half of those to China alone. The loss of these skilled workers, engineers, designers and scientists has undermined our middle class and dangerously eroded our technical, industrial and innovative capacity."

MP: It's true that the U.S. has lost more than 5.5 million manufacturing jobs in the last ten years, from more than 17 million jobs in 2000 to fewer than 12 million jobs in 2010 (see top chart above, data here), which is a 32.5% reduction in factory employment.  And yet during that same period, manufacturing output (data here) actually increased by more than 5%, from $3.1 trillion in 2000 to $3.26 trillion (measured in 2005 dollars) this year (see chart above).  On a per employee basis, manufacturing output per worker increased by more than 50%, from $182,000 in 2000 to $278,000 (measured in 2005 dollars) this year (see bottom chart above). 

Since manufacturing output bottomed in June 2009 at $2.97 trillion due to the recession, there has been a 10% increase in the gross value of industrial output to $3.26 trillion, during a period when manufacturing employment actually decreased by almost 1 percent (and by more than 100,000 jobs), resulting in an 11% increase in output per worker in just a little more than a year.  That surge in manufacturing worker productivity is the largest increase ever over a one-year period, going back to 1972 (see chart), and we can thank the recession for that efficiency surge, as companies were forced to learn how to increase output with fewer workers. 

Bottom Line: Manufacturing worker productivity has doubled in the last 17 years since 1993, and that has contributed to the loss of more than 11 millions jobs. That is, if factory workers were only as productive today as their counterparts in 1993, it would require more than 23 million factory workers today to produce $3.26 trillion worth of manufacturing output today, instead of the 11.7 million workers employed today to produce that much manufacturing output. 

Therefore, it's the significant increase in the manufacturing worker productivity that is more responsible for the loss of millions of U.S. manufacturing jobs than the value of the Chinese currency.  Instead of pressuring China to strengthen its currency to increase American factory jobs, it might be even more effective to pass legislation that would make it illegal to increase worker productivity, e.g. enacting legislation prohibiting the use of labor-saving technologies such as internal-combustion and turbine engines, as Don Boudreaux suggests here.


At 10/04/2010 3:28 AM, Blogger James said...


I am not buying this productivity argument.

If productivity is the cause of the lose of manufacturing jobs then why did it not happen 100 or 200 years ago? Productivity increase has been happening in this country since George Washington started buying gun powder to fight the British. Productivity really took off after the imposition of protective tariffs in 1828. The high cost of labor and protected markets gave rise to an explosion in innovation and labor saving devices such as the mechanical reaper of Cyrus McCormick (1834) and the steel plow of John Deere (1837). That productivity improvement did not cause soaring job loses.

This totally bogus conclusion is the result of the flawed ways that both productivity and manufacturing output are measured. Productivity is considered to increase if the number of man-hours used to make a product is reduced. Regardless of how that man-hour reduction is achieved. The fallacy is that when a component part is outsourced to China the man-hours to make that product go away. That reduces the man-hours applied to make the product and counts as a productivity increase. That is bogus. That should count as man-hours lost to free trade. The value of that outsourced component is still counted in the manufacturing output just as it was before even though it is no longer made here. That too is bogus.

It is free trade not productivity that is the cause of the job loss.

Bad data leads to bad conclusions. Garbage in, garbage out - GIGO.

At 10/04/2010 5:01 AM, Blogger rjs said...

people didnt suddenly become more capable, so youre basically saying that technology and automation cost jobs, right?

At 10/04/2010 5:08 AM, Anonymous Anonymous said...


I work in manufacturing, and I buy the productivity argument because I have the numbers for our place and I doubt we are an anomaly. We now ship as many finished tons of steel daily as we did 12 years ago using 1/3 as many workers. Additionally, all jobs coming into the factory have required to run (RTR) numbers before they come in. In the past, we figured out how many people it would take for the operation after the equipment was set up. It is much easier to build effiency in than it is to engineer it in after it is bolted down.

I don't know enough about foreign currency to discuss that topic, but I know that productivity in manufacturing is way way up. I also find that it is easier to blame someone else by looking outward (Japanese, Chinese, NAFTA, currency . . .) than it is to look inward and solve the real problems.

At 10/04/2010 5:14 AM, Blogger niknaknoo said...

Measuring factory output in dollars is pretty meaningless, it would make more sense to measure the actual amount of goods produced.

As the dollar depreciates in value it will always appear that we are producing more.

At 10/04/2010 5:25 AM, Blogger Jet Beagle said...


Your interpretation of economic history is incorrect, IMO, for several reasons:

1. For a couple of centuries, the industrial revolution made possible goods which had previously been too costly for most consumers. Consumer demand for goods exploded as productivity brought down the prices of what were once luxury goods. But changing demographics caused consumer demand to shift toward services for much of the past century (aging of the population and relative decline of "housewife" as an occupation).

2. The demand for goods in the U.S. continued to grow faster than manufacturing productivity increased, chiefly because the U.S. population was growing so fast. Birth control changed that population growth starting in the 1960s. By the 1980s, the baby bust started to slow the demand for goods in the U.S.

3. The outsourcing of work as varied as computer programming and security services - the specialization of firms - has caused an apparent decline in BLS statistics on manufacturing employment. A computer programmer working for GM is considered a manufacturing employee. A computer programmer working for a software company under contract to GM is considered a service sector employee. So some of the manufacturing employment decline is simply due to specialization of firms in the U.S.

At 10/04/2010 5:34 AM, Blogger Jet Beagle said...


I think you missed the statement that the output of manufacturing is measured in inflation-adjusted, constant dollars.

Measuring output in number of goods produced might be valid for very short periods, such as a few months or even a year. But the goods produced in the U.S. change constantly. Comparing the mix of goods in 1990 to the mix of goods in 2010 is an apples and oranges comparison.

At 10/04/2010 7:19 AM, Blogger Jason said...

Walt, I'm with you that productivity is up huge. In the automotive engineering sector, we are doing more with the same number of people. Although, my observation is we are not doing it better or even as well.

The actual number of engineers has been increasing, domestically and offshore. Unfortunately the pace of technology is faster than worker capability or growth. As a result, I think we need more people to keep up with the feature explosion that is occuring in consumer vehicles. Unfortunately, the price point of vehicles needs to increase so there is sufficient cashflow for that hiring to occur.

Everyone I know in this business is stressed out. Personally, I'm just happy to have a job so I'll take the stress. But I can't speak for the rest of the white devils...

At 10/04/2010 7:55 AM, Blogger juandos said...

I think Walt G's comment regarding productivity is pretty much on target and a lot of it can be laid at the feet of improved engineering as its applied to the machinery in manufacturing...

If one catches some of the National Geographic channel's programs on new or improved factories (from autos to canning to tools) one will see increased production with fewer and fewer people...

"But I can't speak for the rest of the white devils"...:-)

Jason! Jason! Jason!

Now I thought that was funny...

At 10/04/2010 8:44 AM, Blogger Jet Beagle said...

rjs: "so youre basically saying that technology and automation cost jobs, right?"


Technology and automation have been increasing worker productivity for at least three centuries if not longer. More workers have been freed up to do an increasing number of tasks. At the same time, the economy required fewer workers to do the tasks that had been done before. That's called economic progress. The economic output of all workers increases - and standards of living increase - as technology and automation enable each worker to contribute more.

At 10/04/2010 10:17 AM, Anonymous Anonymous said...


I understand what you are saying, and I sense your frustration. Make sure you are using an objective metric to determine that your area of responsibility is not more effective and efficient now than it was in the past. You might be pleasantly surprised what you find.

Incremental effectiveness and efficiency changes over time and additional workloads distort the seat-of-the-pants feeling of how well or poorly you are performing daily. I owe my current job to being able to perform objective performance analyses. When emotions are high and everyone is overworked and stressed out, a clear voice is appreciated by all sides.

At 10/04/2010 11:22 AM, Blogger Buddy R Pacifico said...

If one assumes that producitity is more responsible then Chinese currency control in U.S. job losses then is that the end of discussion? Absolutely not, because for global trade to make sense then some basic rules need to be adhered to. Richard Trumka is saying give U.S. producers the opportunity to compete and prove they are the most productive.

Do not get into union work rules etc., etc. as a side issue because Trumka is not battling producers in this particular argument. People that support global trade should strongly condemn the absence of a market for the yuan. Free the yuan and take away this particular advantage from Chinese producers.

Here are some other keystone changes to support U.S. producers in global trade:

I. Goods and Services Tax (Value Added Tax) but not as an additonal tax.

II. Reduce substantially or eliminate coporate income tax.

III. Elevate to Cabinet level an Export Affairs position.

At 10/04/2010 12:02 PM, Blogger Unknown said...

There's no doubt that automation and technology have greatly influenced the migration of work to LCCs, and that we need fewer people to manufacture as much or more of what we have. That's progress.

But a fundamental flaw is beginning this point with productivity from the BLS (the Fed gets its prod data from BLS to begin with, too). Let's use accurate data to begin this debate and see some truth before lighting off on a flawed premise.

At 10/04/2010 12:16 PM, Blogger Ron H. said...

AJ said...

"But a fundamental flaw is beginning this point with productivity from the BLS (the Fed gets its prod data from BLS to begin with, too). Let's use accurate data to begin this debate and see some truth before lighting off on a flawed premise."

Great idea! Your recommendations for a source of accurate data are welcome. Links please.

At 10/04/2010 12:30 PM, Blogger dan said...

These are not independent variables. When the cost of foreign labor goes down one would expect an increase in domestic productivity. Labor intensive activities would be the first to be shipped overseas or replaced with higher productivity methods of production. Presto, domestic productivity increases.

This however is the inexorable march that is the hallmark of free trade and is the basis for global growth and economic progress. Fluctuation in currency are transient phenomenom that affect the pace of this phenomenom but the trend is unavoidable. The only real way to deal with is what modern economies have always done - adapt.

To adapt means producers must have ready access to capital, reduced regulatory barriers and tax structures that don't penalize risk taking. This is where the policy prescriptions should be targeted.

At 10/04/2010 1:41 PM, Blogger James said...

Walt G,

You missed the point. I do not doubt that productivity has improved. You say you are using one third of the people you used to. Cyrus McCormick’s mechanical reaper was an even greater reduction. The difference is that then as opposed to now labor was shifted from the American farm to the American factory.

Consider a maker of a simple product like nails. Suppose a disgruntled employee gets some investors to open a factory down the street to make nails with half the manpower of the old factory. The new factory is 50 percent more productive than the old one. Now let us say that instead of opening the factory down the street the new enterprise buys the old factory shuts it down and moves the machinery to China. The new operation is 100 percent more productive according to the way productivity is calculated.

My point is that blaming the resulting American job loses on productivity makes no sense. Rather it is another example of the myths, half truths, and spin free traders use to hide the harm done by free trade without actually lying.

At 10/04/2010 2:12 PM, Blogger Jet Beagle said...

James: "The fallacy is that when a component part is outsourced to China the man-hours to make that product go away. That reduces the man-hours applied to make the product and counts as a productivity increase."


The BLS does not include the value of outsourced component inputs in measuring the output of domestic manufacturing. When a component part is outsourced to China, two things occur in the calculation of manufacturing productivity:

1. the value of the component is automatically removed from the numerator of the productivity equation;

2. the labor hours transferred from the U.S. to China are automatically removed from the denominator of the productivity equation.

From the Technical Information About the BLS Major Sector Productivity and Costs Measures:

"Real gross domestic product in the business and nonfarm business sectors as published by the Bureau of Economic Analysis (BEA) of the U.S. Department of Commerce is the basis of the output components of the major sector labor productivity business measures."

For those who are not familiar with the term, "real gross domestic product" is the inflation-adjusted value added to products by operations (factories) located in the U.S.

At 10/04/2010 2:50 PM, Blogger James said...


See a non-government view. For example:

”The fundamental shortcoming is in the way imports are accounted for. A carburetor bought for $50 in China as a component of an American-made car, for example, more often than not shows up in the statistics as if it were the American-made version valued at, say, $100. The failure to distinguish adequately between what is made in America and what is made abroad falsely inflates the gross domestic product, which sums up all value added within the country.

American workers lose their jobs when carburetors they once made are imported instead. The federal data notices the decline in employment but fails to revalue the carburetors or even pinpoint that they are foreign-made. Because it seems as if $100 carburetors are being produced but fewer workers are needed to do so, productivity falsely rises — in the national statistics.

In other words harm from free trade is made to look like it is a productivity increase.

What you showed me is government wishful thinking.

At 10/04/2010 3:43 PM, Blogger Ron H. said...

You missed the point. I do not doubt that productivity has improved. You say you are using one third of the people you used to."

James, it appears to be you who has missed the point. Where do you imagine those other 2/3 of the workers Walt G. mentions have gone? Do you picture them standing in the street outside their former place of employment since their last day of work there? They have likely gotten other jobs, perhaps in the service sector, as there now fewer total manufacturing workers.

Your example of McCormick's invention is a good one. It didn't so much reduce the number of farm workers suddenly, as it did dramatically increase the amount of food being produced, causing prices to fall, so that more people could afford enough to eat, and in fact have more money to spend on other things like manufactured goods.

Continued innovation and improvements in agricultural productivity have caused output to increase, prices to decrease, and the number of workers required to shrink to where it is today with 2-3% of the population working in agriculture rather that 40% as in 1900. Hopefully you see this as a good thing. The same thing is happening in manufacturing, and we shouldn't be alarmed about it.

"The difference is that then as opposed to now labor was shifted from the American farm to the American factory."

Where do you think they go now? Now they shift from the American factory to the American service sector.

Your "nails" story is having trouble with numbers. If the 2nd factory requires 50% as many workers, it is 100% more productive, not 50%. You could also say 200% as productive. The offshored production requires 0% as many workers, so I suppose it is infinitely more productive. In fact, it is no longer an American manufacturer, but an importer. Different measurements. I wouldn't expect someone with exposure to a graduate level statistics course to stumble on something this simple.

At 10/04/2010 4:09 PM, Blogger Buddy R Pacifico said...

Jet Beagle and James, interesting debate on productivity. Jet Beagle's gov't info claims that foreign inputs are taken out of final output numbers. James's private citation says that foreign inputs work their way back into the output figures.

I tend to agree with James and his sources. The guy that attaches the Japanese wing to the Italian fuselage of the 787 at the big B in Everett, WA would have little productivity if all the value was taken out of the two big parts being assembled! 8 hours labor/$0 output (value of foreign sourced wing and fuselage already accounted for using BLS). James is stating that producitivy has had huge gains because: 8 hours worked/joined wing and fuselage output valued in millions (even though most of labor done overseas).

I hope I have stated both Jet and James viewpoints correctly with my example.

At 10/04/2010 6:12 PM, Blogger Jet Beagle said...


Simple reasons why manufacturing GDP as a percent of total GDP has declined:

1. efficiency gains in manufacturing were far greater than efficientcy gains in services (example: robots and machines replaced painters, spot welders, tire installers, and other auto plant workers)

2. our demand for services expanded greatly with government programs and demographic changes(example: medicare combined with the explosive growth in the senior population led to huge increases in medical services)

3. service tasks once done in-house by manufacturing sector firms were outsourced (but not offshored) to firms in the service sector (example: EDS, a service sector company, became the computer programming department for General Motors)

The GDP phenomenon you referenced is not a decline in the manufacturing sector, but rather a huge increase in services, government, and construction sectors.

At 10/04/2010 7:18 PM, Blogger marmico said...

huge increase in services, government, and construction sectors

Private sector services, yes. The secular trend of the increase in services and decrease in goods production has been ongoing since 1947. I believe services overtook goods as a percentage of GDP in 1979.

Government and construction, no.

Since 1947, the private sector's share of the economy has been ~86%. Perhaps you are confounding government consumption and investment with government spending. Government consumption and investment are recorded in the GDP accounts (just like the private sector). Government transfers are not. Perhaps you are not aware of this data tidbit. Federal government consumption and investment was less in 2009 (4.4%) than when St. Ronald assumed the throne in 1981 (5.5%).

Construction as a percent of GDP has averaged ~4.5% since 1947 with a low of 3.9% (1993) and a high of 4.9% (multiple years, the last being 2006).

So there has been little material change in the share of construction and government.

At 10/04/2010 7:27 PM, Blogger juandos said...

"Technology and automation have been increasing worker productivity for at least three centuries if not longer"...

Well Jet Beagle you could tell the tales about how improving and new technology took crew folks out of the cockpit, right?

At one time there was not only the pilot and co-pilot but a flight engineer, a navigator, and a radio operator...

Now there's but two people left...

At 10/04/2010 7:34 PM, Blogger Ron H. said...

Now there's but two people left...

And at least one of them is armed.

At 10/04/2010 7:39 PM, Blogger Jason said...

@Walt: It is really hard to measure productivity in engineering. Much of what we do is wasteful as we generally occupy the gray space between research and production. The bugs that need fixing come at a cost of productivity. And my observation is the integration of technology is occurring too fast for the engineers to keep up. And adding more engineers is not necessarily a solution much, if not most, of the time. I would like to think I am good at what I do, so I consider the increased workload as a sort of "occupational hazard." But I see the majority of people in the industry struggling.

Another observation is that in automotive, as in many industries I imagine, that there is a 20-60-20 rule: 20% are turds, 60% obedient enough, 20% people who get it. YOU CANNOT DO WITHOUT THE TOP 20%. But you need the others to keep the top ones from working themselves to death. And you really have a hard time recruiting above that 20% number - believe me I've tried.

In America we ride that top 20% better than any civilization in history.

At 10/05/2010 10:35 AM, Blogger Jet Beagle said...


Are you intenrtionallty trying to shift the discussion? You initially referred to manufacturing, but in the last post you referred to "goods production". The goods producing sector includes not just manufacrturing but also agriculture, forestry, mining, and construction. Almost all the discussion in this thread has been about manufacturing, not goods production.

In any case, I think you are mistaken. According to the BEA, these were the shares of GDP in 1947:

private services producing - 47.8%

manufacturing - 25.6%

other private goods producing - 14.2%

The private services sector was greater than the goods producing sector as long ago as 63 years, and it was double the manufacturing sector back then. The private services sector has been outgrowing both for the entire period since then.

You are correct that the government sector - as measured by the BEA - has not greatly increased. I was mistakenly thinking about government spending over the 20th century, which did grow explosively. However, the BEA counts most government spending with private industry - such as defense, Medicare, and medicaid - as private industry GDP.

At 10/05/2011 3:39 AM, Blogger OBloodyHell said...

>>> And you really have a hard time recruiting above that 20% number - believe me I've tried.

Trust me, Jason, as someone who is above that 20% number, but remains unemployed, it comes from lazy-ass HR people who want trivial ways to separate the easy hires from the tough hires and stop there.

..."tough hires" being ones they'd actually have to really investigate to vet, and then have some small measure of risk anyway.

HR people want easy paperwork hires -- make a couple phone calls, see that the paperwork matches, and "Hey, I'm done!"

I am one of the best programmers I know, and certainly one of the best software testers there is - I have a very natural feel for where software screws up -- and yet I can't find a freakin' job in the software industry despite 30 years of experience because I don't have the "full degree" (I only took the PROGRAMMING courses, mind you!) and haven't wasted my money getting a lot of certifications that are useless after about 2 years (Either I can't afford to get the certifications because I don't have a job, or I don't need the certifications because I HAVE a job and taking certifications then is a total waste of time if the employer doesn't need a specific one -- because anything you take is pretty much outdated within 2-3 years and useless)

I KNOW I'm very, very good, by a number of objective measures... but trying to even get an interview -- that is, getting past HR to the actual managers and stuff who KNOW what they need and can tell if you've got a clue by talking to you and asking pertinent questions -- is practically impossible.

So, when you say you can't get to the best hires -- part of that is HR isn't looking for them, because they often lie OUTSIDE the standard career advancement protocols.



Post a Comment

<< Home