Thursday, April 26, 2012

Phenomenal Gains in Manufacturing Productivity

Today's factory workers produce more output in an hour than workers in the 1940s produced in a day.
The chart above shows annual real manufacturing output per worker from 1947-2011 using data released today by the BEA for GDP by industry, and data from the BLS on manufacturing employment. In 1950, the average U.S. factory worker produced $19,500 (in 2011 dollars) of output, and by 1976 the amount of output per worker had doubled to $38,500. Output per worker doubled again to $74,400 (in 2011 dollars) by 1997 (21 years later) and then doubled again to $152,800 by 2010, but it only took 13 years for the last doubling because worker productivity has been accelerating.  Last year, manufacturing output per worker increased to a new record high of $156,500 (see chart), and almost ten times the output per worker in 1947.  In other words, the average American factory worker today produces more output in an hour than his or her counterpart produced working almost a ten hour day in 1947 - and that's why we're producing record levels of output with fewer workers.   

This is an amazing story of huge increases in U.S. worker productivity in the manufacturing sector. In fact, the growth in manufacturing worker productivity more than doubled from 2.63% per year in the period between 1950 and mid-1970s to 5.42% annually between 1997 and 2011. Whereas it took 26 years for output per worker to double during the first period (1950-1976), it only took 13 years during the more recent period (1997-2010).

We are constantly inundated with bad news about the decline in the number of manufacturing jobs in the U.S., but we never hear the good news about why that is happening: Manufacturing workers in America keep getting more and more productive, which then allows us to produce more and more output over time, with fewer and fewer workers. That's a great story about an American industry that is healthy, successful and thriving, and not an industry in decline.

By continually increasing worker productivity and productive efficiency, the American manufacturing sector has been hugely successful at achieving one of the most important economic outcomes of being able to "produce more with less." In the process, those efficiency and productivity gains have helped conserve scarce resources, including human resources, more effectively than almost any other industry, except maybe farming. It's hard to overstate how much the efficiency gains achieved by U.S. manufacturing have contributed to the improvements in our standard of living by making manufactured goods more affordable over time. We should spend less time complaining about fewer workers in manufacturing, and more time celebrating the phenomenal gains in manufacturing worker productivity.

Update: The chart below shows annual real manufacturing output and annual manufacturing employment between 1947 and 2011.  Note that manufacturing employment dropped by one-third from 17.56 million jobs in 1998 to 11.52 million in 2010.  


26 Comments:

At 4/26/2012 9:31 PM, Blogger Unknown said...

Yet according to Obama, some must get poorer for others to be wealthier.

 
At 4/26/2012 9:58 PM, Blogger kmg said...

I am eagerly waiting for the time when the sole, last remaining worker produces the full $3 Trillion of output all by himself.

aka The Singularity.

 
At 4/27/2012 5:30 AM, Blogger Larry G said...

the phenomenal gains in productivity - spur businesses to seek alternatives to hiring more employees, first.. especially when the economy is sputtering and fragile.

hardware and software - replace people.

more and more, businesses get down to the essence of what only a human can do and everything else is automated and mechanized.

"Futurists" used to "predict" this would happen and it was said even back then - that it would be problematical when/if a "tipping" point of ratio of production needed vs population was reached, passed.

In short - "excessive" population growth can exceed the economy's ability to accommodate them if from a production perspective, each worker can produce the needs of hundreds of others.

those danged pesky demographics and actuarials mess up nirvana.

 
At 4/27/2012 6:01 AM, Blogger Unknown said...

What kmg said ... if the BLS would revise its formula for measuring US manufacturing productivity, I's be interested.

 
At 4/27/2012 6:07 AM, Anonymous Anonymous said...

It is impressive.

Yet with all the gains in productivity wages have not increased all that much. So while the CEO gets a huge bonus the production workers wage growth remains flat.

And before you bash me as a flaming lib I can assure I am very much a fiscal conservative. If profits and productivity are soaring then wage growth should at least reflect this.

 
At 4/27/2012 6:25 AM, Blogger Larry G said...

if two companies in competition with reach other experience equivalent gains in productivity - they see those gains as opportunity to undercut the other and gain market share.

paying their employees more - actually weakens their competitiveness.

right?

I'm not advocating this, just saying it probably is the reality.

treating your employees "good" first and foremost means operating the company so as it will stay in business and prevail against it's competitors.

the best the company can do is pay the employees what they are worth - to any employer.

companies no longer "take care" of employees.... perhaps never did...

employee's need to be loyal their own best interests FIRST, also.

it's business after all.

 
At 4/27/2012 9:06 AM, Blogger morganovich said...

worth remembering when you look at this graph:

the inflection point in the early 90's coincides with the bls changing the cpi calculation. whichever method you believe to be correct, there si no question that the new cpi reads lower.

that means that for any given set of facts, it will show more productivity gains.

you really need to treat this chart as two grafted series.

 
At 4/27/2012 9:34 AM, Blogger SteveH said...

Then why is the lie about productivity slowing since 1973 allowed to stand unquestioned? Like Tyler Cowen's The Great Stagnation...

 
At 4/27/2012 9:37 AM, Blogger Mark J. Perry said...

The data in the chart have nothing to do with the CPI from the BLS. The data were converted into constant dollars using the GDP Deflator for the manufacturing sector from the BEA.

 
At 4/27/2012 9:52 AM, Blogger Che is dead said...

"Yet with all the gains in productivity wages have not increased all that much. So while the CEO gets a huge bonus the production workers wage growth remains flat." -- Bill

You have to look at total compensation, including health insurance. Once you have done that the "workers wage growth remains flat" myth explodes.

 
At 4/27/2012 10:00 AM, Blogger Che is dead said...

... nearly all reporting on income inequality in America has suggested that the incomes of the rich have been rising, while incomes for the rest of us have been stagnant or even declining. But that may represent a significant misreading of the data.

Most studies of inequality, including the recent widely reported study by the Congressional Budget Office, rely on IRS-reported taxable income. But, as studies by the Cato Institute’s Alan Reynolds and others show, reports of skyrocketing incomes among the top 1 percent of earners may be distorted by changes in the tax code that have resulted in more wealth being reported as taxable income. These tax changes caused businesses to switch from filing under the corporate tax system to filing as individuals, and executives to switch from accepting stock options taxed as capital gains to nonqualified stock options taxed as salaries. Simultaneously, the reductions in income-tax rates in 1986 caused much previously unreported income to show up on tax returns.

At the same time, incomes among lower- and middle-income workers have been shifting from cash wages to non-cash benefits such as health insurance and pensions. These non-cash benefits frequently do not show up as taxable income even though they have value to the worker. In fact, a recent study by Mark Warshawsky of the Social Security Advisory Board suggests that nearly all of the recent increase in earnings inequality “can be explained by the rapid increase in the cost of health insurance employee benefits, and that therefore [there] has not been as significant increase, if any, in inequality of compensation.” -- National Review

 
At 4/27/2012 10:43 AM, Blogger Larry G said...

re: total compensation

Yes, Where is THAT chart?

I suspect that is closer to the truth - that health care costs have eaten up the increased compensation for many.

 
At 4/27/2012 10:44 AM, Blogger SteveH said...

I agree that the BLS productivity numbers show a slowing rate but then a rising rate depending on your choice of lag times. It seems that the rate of change bottomed out in the mid 1980s and increased unevenly from there with a slight decline in 2011. Since productivity has been improving it seems to be disengenuous to always characterize it as a declining rate even into the present day even though the times to doubling output are still longer today than they were from 1947 to 1973 they are shortening again after the mid 1980s.

 
At 4/27/2012 11:23 AM, Blogger Its GSATT said...

Bill "Yet with all the gains in productivity wages have not increased all that much. So while the CEO gets a huge bonus the production workers wage growth remains flat."

He might get a fat bonus, but more of the incentive to make more efficient workers is to keep themselves afloat. EPA slams them with a gismo to purchase to be green and they can either pass that on to the customer, or find another way to make up for the penalty costs. Labor and production costs are the top costs to a company and potentially loooootttssss of room to squeeze some dollars out of.

Good to be efficient, yes. do unions hate it? Bahahahahaha. I bet it's hard to defend an auto plant worker making $20/hr to place a part on a machine a press a button.

 
At 4/27/2012 11:43 AM, Blogger VangelV said...

In short - "excessive" population growth can exceed the economy's ability to accommodate them if from a production perspective, each worker can produce the needs of hundreds of others.

What nonsense. Try and have a chat with a mining or oil company. They are dying for qualified employees willing to work hard. There are $100K plus jobs unfilled because nobody who is qualified wants to apply for them.

The problem is the huge cost of labour thanks to regulations that make it hard to hire and fire unskilled people. That means that low value added jobs are shipped abroad and companies use machinery to do more and more of the work that used to be done by people.

 
At 4/27/2012 12:04 PM, Blogger morganovich said...

mark-

the bea numbers for price level underwent the same changes that the bls ones did.

they began to change in the 80's and underwent dramatic revisions in the early 90's.

the graph of BEA data is still a data splice which makes the slope of increase of the 90's relative to prior periods highly suspect.

 
At 4/27/2012 12:07 PM, Blogger Larry G said...

"... companies use machinery to do more and more of the work that used to be done by people. "

wouldn't that be anywhere in the world where machinery is cheaper than human labor?

why would any company anywhere hire people for more than what machinery would cost for the same output?

is it possible to have more population that we can have jobs for - ever?

 
At 4/27/2012 12:45 PM, Blogger VangelV said...

wouldn't that be anywhere in the world where machinery is cheaper than human labor?

why would any company anywhere hire people for more than what machinery would cost for the same output?

is it possible to have more population that we can have jobs for - ever?


When you replace many people with machines you will see an increase in productivity. The displaced should go where they are most useful but in the case of the US and EU those people have few incentives to look for lower paying jobs that would be the basis of a new career.

 
At 4/27/2012 12:48 PM, Blogger Mark J. Perry said...

The inflection point for productivity in the 1990s happened because that's when manufacturing employment "fell of a cliff," and probably has nothing to do with any revisions in the BEA's calculation of manufacturing deflator.

 
At 4/27/2012 1:03 PM, Anonymous Anonymous said...

Larry,

paying their employees more - actually weakens their competitiveness.

right?


No.

companies no longer "take care" of employees.... perhaps never did...

They take care of their employees to the extant that employees take care of their employer. Employment isn't a one way street. It is a trade. The employee trades his labor for money.

employee's need to be loyal their own best interests FIRST, also.

And yet you're writing this comment as if employers should be loyal to their employees interests first. If an employee didn't like working somewhere, I'm sure you favor the idea that they can quit (for any reason he wants) and looking elsewhere for employment. But you see something sinister in an employer firing someone because they don't like the employee.

Are you unaware that you hire and fire people daily? Have you ever eaten at Outback? Did you today? You fired them for the production of your food today. Are you engaging in some anti-Outback waiter activities by doing this?

 
At 4/27/2012 1:04 PM, Blogger morganovich said...

mark-

if you shave several percentage points off your inflation measure, and i do not believe anyone is arguing that we did not do that, only whether or not it was appropriate, then how could it possibly not affect the productivity numbers?

less reported inflation = more apparent productivity from any given set of real world states.

if boskin et al are right, then perhaps the post change slope is the correct one (though i personally do not believe that) but that also means the slope pre changes is too low.

there is no way to get around the fact that if, for the same facts, you call inflation lower, productivity will read higher. this makes the data inconsistent across that change.

it's not directly comparable. lots of us data now has this problem.

 
At 4/27/2012 1:19 PM, Blogger Larry G said...

" And yet you're writing this comment as if employers should be loyal to their employees interests first"

I'm pretty sure that I did NOT say that and if implied.. my error.

and I thought I made clear that IF there is an employer duty to employees it is a successful company that continues their jobs -FIRST.

and I went on to say that "loyalty" is not now and perhaps never was a good basis for the employer-employee relationship.
( DUH? ).

both have their own interests and both need to be loyal first to those interests.

years ago...back in the day, as they say - people looked to find work with a company that they'd plan to stay with for a career.

they'd buy a house nearby on the premise that the 30 year mortgage was pretty much assured as long as they did a good job.

much of that is gone now for many people and an employee is "worth" only what other employers are willing to pay for him/her.. not what your previous employer once valued you at ...

I've heard that some employers now say if you are currently unemployed, don't apply.

 
At 4/27/2012 1:22 PM, Blogger Larry G said...

" When you replace many people with machines you will see an increase in productivity. The displaced should go where they are most useful but in the case of the US and EU those people have few incentives to look for lower paying jobs that would be the basis of a new career."

that's an intriguing comment and perhaps I misunderstand it.

what does that mean in general as productivity continues to increase with regard to available jobs - even the ones that require increased skills/education?

is it just a mismatch in needed skills or is work harder and harder to find?

 
At 4/27/2012 1:29 PM, Anonymous Anonymous said...

This comment has been removed by the author.

 
At 4/27/2012 1:30 PM, Anonymous Anonymous said...

I'm pretty sure that I did NOT say that and if implied.. my error.

I thought you were getting at when you said "companies no longer "take care" of employees". Thanks for clearing that up.

both have their own interests and both need to be loyal first to those interests.

Agreed. I think it's important that workers keep this in mind. Many of the most vocal don't and come off looking like self-entitled assholes.

much of that is gone now for many people and an employee is "worth" only what other employers are willing to pay for him/her.. not what your previous employer once valued you at ...

I think employers have always paid an employee what his market value was, which is a combination of "what other employers are willing to pay" and what the employer valued you at. I'm not sure how much previous pay ever played in determining current salary. I'm sure some, but I'm willing to bet not as much as most think.

 
At 4/27/2012 6:36 PM, Blogger VangelV said...

This comment has been removed by a blog administrator.

 

Post a Comment

<< Home