Productivity has increased massively. What the hell does that mean

Factual Questions
1

This has been reported so often that’s it’s practically become a meme – “US Productivity since (fill in the year) has increased by a massive amount. we have simply become incredibly efficient.”
The statement is used as an arguing point by both sides – pro-business types argue that it shows that US manufacturing is still strong, despite the appearance that factories are closing and that people are saying that manufacturing is going overseas – we’re as good as ever, we’re just more efficuient at it. Pro-worker folks argue that efficiency is up along with profits, but that worker’s paychecks don’t reflect this. the gains due to increased productivity are going to the managers, and they say that ain’t right.

I am, myself, suspicious of the premise. What does it mean to say that productivity is up? It’s not as if there was some massive plague of inefficiency twenty or thirty or even fifty years ago. they had efficiency experts working even then to make things work well. i have observed manufacturing through my own jobs and that of my relatives throughout my lifetime, and i simply don’t see the massive gains in producticvity that people claim are there. Plastic injection molding in the 2000s looks remarkably like plastic injection molding in the 1960s. There certainly hasn’t been an order-of-magnitude (or even simply a large) increase in production speed, or in percentage of parts passing inspection, or in effective use of workers’ time. And this has, in general, been true for all the manufacturing industries I have any contact with.

So what the heck does this statement mean? I can suggest some possibilities:

1.) It is used for industry as a whole, so that, while many types of manufacturing have seen no increase in efficiency, some others have, and that brings the average up.

2.) Improvements in technology in some areas has made productivity, measured in certain ways, rise. So Moore’s Law of increasing computing speed/storage volume/whatever has caused massive improvements in “computer productivity” , again bringing up the average when you factor it in along things like baby-carriage manufacturing.

3.) Improvements in warehousing (especially its automation) and managing inventory is counted as a major contributor to productivity growth (same reference as above).

4.) a lot of U.S. manufacturing is “high end” products using components largely manufactured elsewhere, so measured in the price of finished units, productivity has gone up, but it means not that we’re more efficient at making stuff altogether, but are choosing to make relatively few very expensive things from parts we buy cheaply elsewhere.

5.) Moving not only parts manufacturing but also service providers offshore has resulted in increases in measured US productivity ( Economist's View: Increased U.S. Productivity from the Outsourcing of Services )

and so on.

my point isn’t to condemn any of these actions. My point is that the bald statement that US Productivity has Increased is vague (I still have no idea how it is calculated, or what it really means) and tremendously misleading. It doesn’t seem to mean what most people seem to assume it means, from what I can gather. It gives the impression that we;re practically turning raw materials into finished products just by looking at them, when I suspect it means that we’re losing less money by efficiently storing and retrieving things, finally getting around to automating stuff that could’ve been automated a long time ago, and cutting costs in new ways. But the machines are pretty much ka-chunking away in the same way and at the same rate, only we have a few more robots than before, and we’re not making as many low-end products. But it doesn’t mean that we’re wonderful manufacturers of evetrything, or that there’s more wealth per widget to be spread around.
Comments?

2

You’re going to be able to find a lot of definitions and arguments for what “productivity” means because the truth is it means whatever the writer wants it to.

In general, “productivity” means the amount of output generated per unit of X. There are, consequently, a variety of different potential definitions of productivity:

  • Labor productivity, which can be measured by output per worker, per worker-hour, or whatever other way you want to measure labor

  • Capital productivity, e.g. prodictivity per unit of capital, or capital inputs

  • A zillion others

I suspect if you were to comb through “productivity” related claims from 100 sources you’d find them using 100 different definitions; there is no agreed-upon central definition. There really can’t be, in fact.

Obviously, fixing productivity to output versus a single variable allows the speaker to make almost any claim he or she wants irrespective of reality, because you can cherry pick the measure you want to use.

This isn’t to say that the concept isn’t useful. Any well run business analyzes productivity in hundred ways to find how better to make money. I was at a business with 50 employees yesterday and they measured productivity in more ways than they had employees (if you mixed the variables up a little.) But be suspicious of anyone using a single productivity measure as support for an argument.

3

Generally I hear productivity used as a shorthand for labour productivity, that is, the number of units of goods produced per unit of labour input. In more accessible terms, productivity is a measure of how many goods are produced for each person-hour worked.

4

This is how I would interpret reports of growing productivity. If reports would refer to one of the others, they’d be more likely to add ‘capital’ to their story. The message of these reports is mainly: we work just as long, but simply produce more than we used to*.
*which is off course aided by machines that fall under capital, but that’s beside the point.

5

It’s funny that I see the exact opposite of your observations. Every manufacturing process I’ve seen first hand has had massive productivity improvements: meat processing, dairy processing, factory manufacturing cosmetics, building of mass-produced guitars.

Robots, CNC, lasers, computer automation, improved materials, faster curing times of re-formulated plastics and paint, more streamlined management of work materials flowing between humans on the factory floor, computation of optimal production schedules making better use of the 3rd shift from the minds of mathematics PhDs , etc. Every little thing adds up.

For example, the subject of growing corn in USA has been a media favorite for the last couple of years. There must have been a dozen documentaries about it from Discovery Channel, National Geographic, History, etc. Several productivity leaps stick out: more corn yield per acre, less human labour to harvest, less transporation time from the field to the plate of food in your home, etc. Corn is not unique with its story of productivity. The leaps of technology it uses has overlaps with every other non-food related industry. Productivity improvements are everywhere.

6

I think the tremendous productivity of automated manufacturing spells doom for a lot of labor. Take a steel mill-a modern mill has a few hundred employees, rather than the thousands needed in an old fashioned plant. Automobile manufacturing: the direct labor has been reduced beyond henry ford’s wildest dreams.
Paradoxically, intangible services (like leagal services) are curiously immune to productivity increases-indeed, they cost more and are less efficient, as time goes on (most state courts have resisted computerization-it is bad for jobs.)

7

Consider banking.

Imagine the productivity increases in that field since before WWII.

I don’t even know if it would be possible to run the banking systems of most G8 countries using rows and rows of clerks.

8

In the manufacturing of new items, there is certainly a lot to be gained through experience at the start of the cycle. Lasers are pretty new technology, and in the early years, were built entirely by hand in a laborious process. Certainly that could be speeded up, especially when economies of scale come in. But many lasers I’ve seen did not see such an improvement. the Color Center lasers I saw manufactured in upstate New York still took a long time per unit. And it’s not fair to compare them to mass-produced semiconductor lasers that are run off by the same sort of techniques used by semiconductor devices – it’s comparing apples and oranges.

as I’ve said, injection-molded plastic items still used essentially the same processes and speed as in the 1960s, despite improvements in many places. So did processing of absorbent pads used in diapers, sanitary napkins, and the like. I’ve seen the actual production floors for these over decades.
So, yes, you do get things like ultra-fast UV cure for paints, but that only affects certain industries, and only parts of those. My point is that productivity in many items hasn’t notably improved – they’re still done the same way, and at the same rate. I suspect that the gains in speed for some items (and the speed efficiency of just-in-time, rapid delivery, and other such management things) are making the average gains faster.
Nor do i think these are fringe or atypical industries. Pill-making machines and bottling of medicines isn’t noticeably different today than 30 years ago. Nor soda bottling, nor container manufacture (glass bottles, plastic bottles, etc.).

9

In essence, productivity is measured by dividing the value of output (manufactured goods or services rendered) by hours worked.

So productivity is increased if more output is produced from the same amount of labor, or if less labor produces the same amount of output.

10

It’s not exactly that we’re producing things much faster nowadays. It’s that we’re producing things a bit faster, but with a lot fewer workers. If you have a production line where 100 people screw the tops onto beer bottles, and replace that with an machine that screws the tops on and have just 4 guys maintaining and monitoring the machine, that’s a 25x gain in productivity even if the machine works at exactly the same speed and gives the exact same results as the human workers did. Given the new capital improvement 4 workers can produce the same amount as 100 workers.

And US manufacturing, measured in dollars, is higher than ever, and is much larger than the Chinese manufacturing sector. Domestic manufacturing isn’t disappearing, domestic manufacturing jobs are disappearing because modern factories need a fraction of the workers.

11

Right. Also, US tends to manufacture expensive non-consumer items. I was just at my doctor’s office the other day. He works in a mordern, large medical group. I happened to check quite a few of the medical devices as I passed by: the instruments, the tables, the scales, etc. All were made in the U.S.

12

Is this, indeed, the technical definition? That’s my question. I suspect that not everyone is using the same definition.

Although i see this sort of thing in robot-aided industries, my own experience is that there hasn’t been the massive increase in automation that people suggest. In places I’ve worked, there have been a minimal number of people running the production machines, but that was the case twenty, thirty, and more years ago. Automation has been a fact of life in manufacturing in a big was since the 1950s. I could imagine large gains there, but I’m suspicious of large gains in the past decade or so.

13

That’s labour productivity.

As others have mentioned there is also capital productivity.

14

I keep hearing this, but I don’t see it. Every time I see a manufacturing floor, it loooks like the samne sort of mechanized factory I saw thirty years ago. I don’t see banks of roboticized assembly lines presided over by one or two workers. I see the same sorts of automated production lines supervised by the same number of workers I saw back then. It’s not as if we now have almost total automation of stuff that was done by uinskilled hand labor thirty years ago – this stuff was automated already thirty years ago.

15

Here’s an interesting theory that productivity in the 1990s rose because of massive layoffs of middle-management workers:

16

It isn’t necessarily automation. Productivity increases when the same number of people increase output. It can be better tools, less waste, better parts requiring less finishing work, fewer mistakes, etc… And, yes, most of the automation happened decades ago. Increasing the automation can decrease productivity in financial terms from higher maintenance costs, higher wages to the remaining personnel who need higher skill levels, plus the fixed cost of equipment that can’t be laid off when demand decreases. Add to that accounting/tax rules that allow costs to be counted in future years after the CEO has moved on, giving only the illusion of increased productivity.

Edit: Your next post is an important point to. Cutting out unproductive/unnecessary costs. But I hear that the price of buying politicians is increasing. Lots of factors in this equation.

17

This seems so obvious to me I don’t even understand why it’s worth mentioning. Of course some industries see productivity improvements at different times than others. Who would expect otherwise?

18

qpw3141 is correct, I meant labor productivity.

From the Bureau of Labor Statistics:

“Productivity is measured by comparing the amount of goods and services produced with the inputs which were used in production. Labor productivity is the ratio of the output of goods and services to the labor hours devoted to the production of that output.”

Cite:

19

The biggest reason its being mentioned and discussed is as the economy improves and US businesses increase production and sales, don’t expect employment to rise as would have been expected before these productivity gains. Which will have a dampening effect on the improvements to the US economy.

20

it also means that components and products that are labor intensive to make are no longer produced in America and hence they don’t affect American productivity numbers while dragging down foreign ones. If our clothes are sewn by people in Pakistan using the current not-very-automated sweatshop technology, it is the Pakistani businesses and workers who end up looking “unproductive” and not American ones.